CN114723468A - Information push method, device, computer equipment and storage medium - Google Patents

Abstract

The application relates to an information pushing method, an information pushing device, computer equipment and a storage medium. The method comprises the following steps: acquiring original consumption resources and candidate resource values corresponding to the target information, and determining a prediction adjustment coefficient corresponding to the candidate resource values; acquiring the converted amount and the accumulated consumption resource of the target information in a first preset time length, and determining the adjustment expected accumulated consumption resource and the expected converted amount of the target information in a second preset time length based on the predicted adjustment coefficient and the accumulated consumption resource; determining a target consumption resource of the target information in the target duration according to the converted amount, the accumulated consumption resource, the expected converted amount and the adjustment expected accumulated consumption resource; the target time length comprises a second preset time length and a first preset time length; and when the target consumed resource and the original consumed resource accord with the information pushing condition, pushing the target information according to the candidate resource value. By adopting the method, the accurate pushing of the information can be realized.

Description

Information push method, device, computer equipment and storage medium

technical field

The present application relates to the field of computer technology, and in particular, to an information push method, apparatus, computer device and storage medium.

Background technique

With the development of computer technology, information push has become an important channel for people to obtain new information in their work and life. If the user is interested in the pushed information, he usually responds to the information to realize the conversion of the information, so as to achieve the purpose of information push. The conversion of information refers to a user's specific operation on the information, such as clicking on the information, downloading the application program in the information, and so on. The conversion amount of push information is closely related to the push method of information, and the amount of information conversion brought by different push methods is not the same.

However, traditional information push methods are often customized, such as the time and location of the push, which results in inaccurate information push.

SUMMARY OF THE INVENTION

Based on this, it is necessary to provide an information push method, device, computer device and storage medium capable of accurately pushing information to address the above technical problems.

An information push method, the method includes:

Obtain the original consumption resource and the candidate resource value corresponding to the target information, and determine the prediction adjustment coefficient corresponding to the candidate resource value;

Obtain the converted amount and accumulated consumption resources of the target information within a first preset time period, and determine the adjustment of the target information within a second preset time period based on the predicted adjustment coefficient and the accumulated consumption resources Expected cumulative consumption of resources and expected conversion;

According to the converted amount, the accumulated consumption resources, the expected conversion amount, and the adjusted expected accumulated consumption resources, determine the target consumption resources of the target information within the target duration; the target duration includes the first 2. a preset duration and the first preset duration;

When the target consumption resource and the original consumption resource meet the information push condition, the target information is pushed according to the candidate resource value.

In one embodiment, the determining, based on the predicted adjustment coefficient and the unadjusted expected consumption resource, determines the adjustment expected consumption resource corresponding to the second preset time period when the target information is adjusted, including:

Determining the product of the predicted adjustment coefficient and the unadjusted expected consumption resource as the adjustment expected consumption resource corresponding to the second preset time period when the target information is adjusted;

Determining the expected conversion amount of the target information within the second preset time period according to the adjusted expected accumulated consumption resources and the adjusted expected consumption resources includes:

The ratio of the adjusted expected accumulated consumption resource to the adjusted expected consumption resource is determined as the expected conversion amount of the target information within the second preset time period.

In one embodiment, the obtaining the historical resource value used for pushing the target information within the first preset time period includes:

acquiring the historical resource values used for pushing the target information within the first preset duration;

The determining the historical adjustment coefficient of the target information within the first preset time period according to the historical resource value and the original consumption resource includes:

Perform a weighted average calculation on the historical resource values to obtain the historical resource average value;

The ratio of the historical resource mean value to the original consumed resource is determined as the historical adjustment coefficient of the target information within the first preset time period.

In one embodiment, according to the original consumption resource, the unadjusted expected consumption resource, the smoothing factor, the predicted adjustment coefficient, and the unadjusted expected accumulated consumption resource corresponding to the first preset duration, Determining the adjustment expected resource consumption corresponding to the second preset time period when the target information is adjusted, including:

Determine the predicted adjustment coefficient, the first product of the unadjusted expected accumulated consumption resource corresponding to the first preset duration and the unadjusted expected consumption resource, and the predicted adjustment coefficient, the smoothing factor, the unadjusted expected consumption resource, and the adjusting the second product of the expected consumption resource and the original consumption resource;

determining the sum of the third product of the smoothing factor and the unadjusted expected consumption resource and the unadjusted expected accumulated consumption resource corresponding to the first preset duration;

Multiply the first product by the second product, and divide by the third product and the sum of the unadjusted expected cumulative consumption resources corresponding to the first preset duration to obtain the target information when the target information is adjusted. The corresponding adjustment expected resource consumption within the second preset time period.

In one embodiment, the target consumption resource of the target information within a target duration is determined according to the converted amount, the accumulated consumption resource, the expected conversion amount, and the adjusted expected cumulative consumption resource, include:

Determine the sum of the accumulated consumption resource and the adjusted expected cumulative consumption resource, and the sum of the converted amount and the expected conversion amount;

The ratio of the sum of the accumulated consumption resources to the adjusted expected accumulated consumption resources, and the sum of the converted amount and the expected conversion amount is determined as the target consumption resources of the target information within the target duration.

In one embodiment, the first preset duration is the duration from the starting point of the target duration to the current moment, the second preset duration is the duration from the current moment to the end of the target duration, and the The target duration is a periodic fixed duration.

An information push device, the device includes:

an acquisition module, configured to acquire the original consumption resource and the candidate resource value corresponding to the target information, and determine the prediction adjustment coefficient corresponding to the candidate resource value;

The first determination module is configured to obtain the converted amount and the accumulated consumption resources of the target information within a first preset time period, and determine that the target information is in the first 2. Adjustment of expected cumulative consumption of resources and expected conversion within a preset time period;

a second determining module, configured to determine the target consumption resource of the target information within the target duration according to the converted amount, the accumulated consumption resource, the expected conversion amount and the adjusted expected cumulative consumption resource; The target duration includes the second preset duration and the first preset duration;

A push module, configured to push the target information according to the candidate resource value when the target consumption resource and the original consumption resource meet the information push condition.

A computer device includes a memory and a processor, the memory stores a computer program, and the processor implements the following steps when executing the computer program:

Obtain the original consumption resource and the candidate resource value corresponding to the target information, and determine the prediction adjustment coefficient corresponding to the candidate resource value;

Obtain the converted amount and accumulated consumption resources of the target information within a first preset time period, and determine the adjustment of the target information within a second preset time period based on the predicted adjustment coefficient and the accumulated consumption resources Expected cumulative consumption of resources and expected conversion;

According to the converted amount, the accumulated consumption resources, the expected conversion amount, and the adjusted expected accumulated consumption resources, determine the target consumption resources of the target information within the target duration; the target duration includes the first 2. a preset duration and the first preset duration;

When the target consumption resource and the original consumption resource meet the information push condition, the target information is pushed according to the candidate resource value.

A computer-readable storage medium on which a computer program is stored, and when the computer program is executed by a processor, the following steps are implemented:

Obtain the original consumption resource and the candidate resource value corresponding to the target information, and determine the prediction adjustment coefficient corresponding to the candidate resource value;

Obtain the converted amount and accumulated consumption resources of the target information within a first preset time period, and determine the adjustment of the target information within a second preset time period based on the predicted adjustment coefficient and the accumulated consumption resources Expected cumulative consumption of resources and expected conversion;

According to the converted amount, the accumulated consumption resources, the expected conversion amount, and the adjusted expected accumulated consumption resources, determine the target consumption resources of the target information within the target duration; the target duration includes the first 2. a preset duration and the first preset duration;

When the target consumption resource and the original consumption resource meet the information push condition, the target information is pushed according to the candidate resource value.

The above-mentioned information push method, device, computer equipment and storage medium determine the adjustment coefficient corresponding to the candidate resource value by acquiring the original consumption resource corresponding to the target information and the candidate resource value. According to the converted amount and accumulated consumption resources of the target information in the past time period, and the predicted adjustment coefficient, the expected conversion amount and the expected accumulated consumption resources of the target information in the future time period are accurately calculated. Fully considering the influence of various factors such as the conversion volume and accumulated consumption resources in the past time period, the expected conversion volume in the future and the expected cumulative consumption resources, etc., it is possible to further calculate the target information in the past time period and the future time period. The average consumption of resources in the total time period to determine whether the average consumption of resources is eligible. When the average consumption resources and the original consumption resources meet the information push conditions, it means that the cost of obtaining and converting the predicted information in the future time period is roughly the same as the cost defined by the information owner, then the target information is pushed according to the value of the candidate resources. Accurate information push can be achieved, and a higher amount of information conversion can be obtained under the condition that the cost provided by the information owner can be met, and the benefit of information push can be improved.

Description of drawings

Fig. 1 is the application environment diagram of the information push method in one embodiment;

2 is a schematic flowchart of a method for pushing information in an embodiment;

3 is a schematic flow chart of determining the adjustment of the target information within the second preset time period based on the predicted adjustment coefficient and the accumulated consumption resources in one embodiment;

4 is a schematic flowchart of an embodiment of determining the corresponding adjustment expected accumulated consumption resources within a second preset time period when the target information is adjusted based on the accumulated consumption resources;

5 is a schematic diagram of an interface for placing a target advertisement on an information flow page in one embodiment;

6 is a schematic diagram of an interface for placing a target advertisement on an advertisement display screen in one embodiment;

7 is a schematic diagram of the principle of an information push method in an embodiment;

8 is a schematic flowchart of an application scenario in which an information push method is applied to advertisement placement in another embodiment;

9 is a structural block diagram of an information push device in one embodiment;

Figure 10 is a diagram of the internal structure of a computer device in one embodiment.

Detailed ways

In order to make the purpose, technical solutions and advantages of the present application more clearly understood, the present application will be described in further detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application, but not to limit the present application.

The solutions provided in the embodiments of the present application may involve technologies such as artificial intelligence (Artificial Intelligence, referred to as AI), machine learning (Machine Learning, referred to as ML for short). Artificial intelligence is a theory, technology and application system that uses digital computers or machines controlled by digital computers to simulate, extend and expand human intelligence, perceive the environment, acquire knowledge and use knowledge to obtain the best results. In other words, artificial intelligence is a comprehensive technique of computer science that attempts to understand the essence of intelligence and produce a new kind of intelligent machine that can respond in a similar way to human intelligence. Artificial intelligence is to study the design principles and implementation methods of various intelligent machines, so that the machines have the functions of perception, reasoning and decision-making. Machine learning is a multi-domain interdisciplinary subject involving probability theory, statistics, approximation theory, convex analysis, algorithm complexity theory and other disciplines. It specializes in how computers simulate or realize human learning behaviors to acquire new knowledge or skills, and to reorganize existing knowledge structures to continuously improve their performance. Machine learning is the core of artificial intelligence and the fundamental way to make computers intelligent, and its applications are in all fields of artificial intelligence. Based on technologies such as artificial intelligence and machine learning, it is possible to obtain the original consumption resources and candidate resource values corresponding to the target information, and determine the prediction adjustment coefficient corresponding to the candidate resource values; obtain the converted amount and accumulated target information within the first preset time period. Consuming resources, based on the predicted adjustment coefficient and the accumulated consumption resources, determine the adjustment expected accumulated consumption resources and expected conversion amount of the target information within the second preset time period; according to the converted amount, accumulated consumption resources, expected conversion amount and adjustment expectations Accumulate consumption resources, and determine the target consumption resources of the target information within the target duration; the target duration includes the second preset duration and the first preset duration; when the target consumption resources and the original consumption resources meet the information push conditions, according to the candidate resource value Push target information, so as to effectively achieve accurate push of target information.

The information push method provided by this application can be applied to the application environment shown in FIG. 1 . The application environment provides computer equipment, such as a server 110 , a terminal 120 , a terminal 130 , and a terminal 140 . The information push method can be applied to the terminal 120, the terminal 130, the terminal 140, or the server 110, and can also be applied to a system including a terminal and a server, and is implemented through interaction between the terminal and the server. For example, the server 110 obtains the original consumption resource and the candidate resource value corresponding to the target information, and determines the prediction adjustment coefficient corresponding to the candidate resource value. The server 110 obtains the converted amount and the accumulated consumption resources of the target information within the first preset time period, and determines the adjustment expected accumulated consumption resources and the expected accumulated consumption resources of the target information within the second preset time period based on the predicted adjustment coefficient and the accumulated consumption resources conversion volume. The server 110 determines the target consumption resource of the target information within the target duration according to the converted amount, the accumulated consumption resource, the expected conversion amount and the adjusted expected cumulative consumption resource; the target duration includes the second preset duration and the first preset duration. When the target consumption resource and the original consumption resource meet the information push conditions, the server 110 respectively pushes the target information to the terminal 120, the terminal 130 and the terminal 140 through the network according to the value of the candidate resource. The terminal 120, the terminal 130, and the terminal 140 receive the target information, and automatically display the target information, or display the target information based on user operations. The terminals can be, but are not limited to, various personal computers, notebook computers, smart phones, tablet computers, and portable wearable devices, for example, the terminal 120 is a smart phone, the terminal 130 is a notebook, and the terminal 140 is a portable wearable device. The server 110 may be implemented by an independent server or a server cluster composed of multiple servers. The terminal 120, the terminal 130, the terminal 140 and the server 110 may be directly or indirectly connected through wired or wireless communication, for example, through a network connection.

In one embodiment, any one of the terminal 120 and the terminal 130 can individually push information with the server 110 . For example, the terminal 120 obtains the original consumption resource and the candidate resource value corresponding to the target information, and determines the prediction adjustment coefficient corresponding to the candidate resource value. The terminal 120 obtains the converted amount and the accumulated consumption resources of the target information within the first preset time period, and determines the adjustment expected accumulated consumption resources and the expected accumulated consumption resources of the target information within the second preset time period based on the predicted adjustment coefficient and the accumulated consumption resources. conversion volume. The terminal 120 determines the target consumption resource of the target information within the target duration according to the converted amount, the accumulated consumption resource, the expected conversion amount and the adjusted expected cumulative consumption resource. The target duration includes the second preset duration and the first preset duration. When the target consumption resource and the original consumption resource meet the information push conditions, the terminal 120 sends an information push instruction to the server 110, and the server 110 pushes the target information according to the candidate resource value in response to the information push instruction.

In one embodiment, as shown in FIG. 2 , a method for pushing information is provided, which is described by taking the method applied to the computer device in FIG. 1 as an example, including the following steps:

Step 202: Obtain the original consumption resource and the candidate resource value corresponding to the target information, and determine the prediction adjustment coefficient corresponding to the candidate resource value.

The original consumption resource refers to the resource set by the information owner to obtain a conversion of the target information. Conversion refers to a user's specific operation on information, such as clicking or browsing the information; when the information represents the download address of the application, the conversion can be the user's download and activation of the application in the information, but is not limited to this. For example, the information owner is the advertiser, the target information is the target advertisement, and the conversion means that the user downloads the application in the target advertisement, and the original consumption resource is the cost that the advertiser is willing to pay for one download of the application.

The candidate resource value refers to the candidate resource value required to obtain the push opportunity for the target information. The candidate resource value may be preset and selected from a candidate set, and there are multiple candidate consumption resources in the candidate set. For example, when the target information is a target advertisement, the candidate resource value refers to a candidate bid given for the target advertisement to obtain exposure. Further, the candidate resource value refers to the candidate bid given for the target advertisement to obtain 1 exposure, such as 3 yuan and 5 yuan.

The predicted adjustment coefficient refers to an adjustment coefficient used to adjust the candidate resource value of the target information. The prediction adjustment coefficient and the candidate resource value have a mapping relationship, and the prediction adjustment coefficient uniquely corresponding to the candidate resource value can be determined through the mapping relationship.

In this embodiment, the computer device can obtain the resources that are set by the information owner and consumed for obtaining one conversion of the target information, that is, the original consumption resources, and obtain any predefined candidate resource value. The computer device determines the prediction adjustment coefficient corresponding to the selected candidate resource value according to the mapping relationship between the candidate resource value and the original consumed resource.

Step 204: Obtain the converted amount and the accumulated consumption resources of the target information within the first preset time period, and determine the adjustment expected accumulated consumption resources and the accumulated consumption resources of the target information within the second preset time period based on the predicted adjustment coefficient and the accumulated consumption resources. Expected conversion volume.

The converted amount refers to the number or times that the target information has been converted by the user within the first preset time period. The accumulated consumed resources refer to the consumed resources accumulated when the target information is pushed within the first preset time period. For example, the accumulated consumed resources refer to the total cost accumulated for the target advertisement has been pushed for 12 hours.

The first preset time period refers to the past time period, and the second preset time period refers to the future time period. Further, the end moment of the first preset duration is the start moment of the second preset duration. For example, the first preset duration is 00:00-12:00, and the second preset duration is 12:00-24:00.

Adjusting the expected cumulative consumption resources means that when the candidate resource value of the target information is adjusted, the expected cumulative consumption resources are pushed within the second preset time period. For example, adjusting the expected cumulative consumption resources means that the target advertisement is pushed in the next 12 hours. The total accumulated cost.

The expected conversion volume refers to the number or times of expected target information to be converted by the user within the second preset time period. For example, the expected conversion volume refers to the expected number or times of conversions on the target ad by users in the next 12 hours.

In this embodiment, the computer device may acquire the converted amount and the accumulated consumed resources of the target information within the first preset time period. The computer device determines, based on the accumulated consumption resources, the corresponding adjustment expected accumulated consumption resources within the second preset time period when the target information is adjusted. The computer device determines the expected conversion amount of the target information within the second preset time period based on the adjustment of the expected cumulative consumption resources and the predicted adjustment coefficient.

Step 206, according to the converted amount, the accumulated consumption resources, the expected conversion amount and the adjusted expected accumulated consumption resources, determine the target consumption resources of the target information within the target duration; the target duration includes the second preset duration and the first preset duration .

The target consumption resource refers to the resource expected to be consumed by the target information to obtain a conversion within the target duration under the condition of adjusting the candidate resource value of the target information. That is, the target resource consumption refers to the resource that is expected to be consumed by the target information to obtain a conversion within the target time period when the target information is pushed using the selected candidate resource value within the second preset time period. For example, the candidate bid for the selected target advertisement is 3 yuan, and if the target advertisement is pushed within the next 12 hours according to 3 yuan, the target consumption resource refers to the cost of obtaining one conversion of the target advertisement within 24 hours.

The target duration includes the adjacent second preset duration and the first preset duration. Further, the end moment of the first preset first preset duration is the start moment of the second preset duration, the start moment of the first preset duration is the start moment of the target duration, and the end of the second preset duration The moment is the end moment of the target duration.

In this embodiment, the computer device may determine the sum of the accumulated consumption resources of the target information within the target duration according to the accumulated consumption resources and the adjusted expected accumulated consumption resources. The computer device may determine the sum of the conversion amount of the target information within the target time period according to the converted amount and the expected conversion amount. The computer device can determine the target resource consumption of the target information within the target duration according to the sum of the accumulated resources consumed by the target information within the target duration and the sum of the conversion amounts of the target information within the target duration.

Step 208, when the target consumption resource and the original consumption resource meet the information push condition, push target information according to the value of the candidate resource.

Information push conditions refer to preset conditions for information push. The information push may be that the target consumption resource is equal to the original consumption resource, and the difference between the target consumption resource and the original consumption resource is less than a threshold, but is not limited to this, and can be set according to requirements.

In this embodiment, the computer device calculates the target consumption resource based on the selected candidate resource value, and when the target consumption resource and the original consumption resource meet the information push conditions, the computer device pushes the target according to the candidate resource value within the second preset time period information.

Further, when the target consumption resource is equal to the original consumption resource, the computer device pushes the target information within the second preset time period according to the value of the candidate resource.

In one embodiment, when the difference between the target consumption resource and the original consumption resource is less than or equal to a threshold, the computer device pushes the target information within a second preset time period according to the candidate resource value.

In the above information push method, the adjustment coefficient corresponding to the candidate resource value is determined by acquiring the original consumption resource corresponding to the target information and the candidate resource value. According to the converted amount and accumulated consumption resources of the target information in the past time period, and the predicted adjustment coefficient, the expected conversion amount and the expected accumulated consumption resources of the target information in the future time period are accurately calculated. According to the conversion amount and accumulated consumption resources in the past time period, the expected conversion amount in the future and the expected cumulative consumption resources, the average consumption resources of the target information in the total time period including the past time period and the future time period can be further calculated, Thereby, it is determined whether the average consumption resource meets the conditions, so as to perform the corresponding operation. When the average consumption resources and the original consumption resources meet the information push conditions, it means that the value of the candidate resource is the most suitable resource value for pushing the target information, then the target information is pushed according to the value of the candidate resource, which can achieve accurate information push, and at the same time A higher amount of information conversion can be obtained at the same cost.

In the traditional way, the information is pushed according to the custom push method, and the conversion volume and conversion cost in different time periods are not considered. In this embodiment, the conversion volume and cost of information by users in the past time period are considered, as well as the estimated conversion volume and cost of information by users in the future time period. Based on the influence of various factors, more accurate predictions can be made. The cost of obtaining and converting information in the future time period, so as to more accurately determine the price of information push and realize accurate push of information. At the same time, the target consumption resources and the original consumption resources meet the information push conditions, indicating that the cost of obtaining and transforming the predicted information in the future time period is roughly the same as the cost defined by the information owner, thus satisfying the cost provided by the information owner. Under the condition of higher information conversion, improve the efficiency of information push.

In one embodiment, obtaining the corresponding original consumption resource and candidate resource value of the target information, and determining the prediction adjustment coefficient corresponding to the candidate resource value, including:

Obtain the original consumption resource corresponding to the target information, and select the candidate resource value for pushing the target information from the candidate set; determine the prediction adjustment coefficient corresponding to the candidate resource value according to the candidate resource value and the original consumption resource.

The computer device may obtain the original consumption resources set by the information owner for the target information, and obtain the candidate set corresponding to the target information. There are multiple candidate resource values defined for the push target information in the candidate set, and the multiple values refer to at least two. The computer device may select a candidate resource value from the candidate set, and calculate a prediction adjustment coefficient corresponding to the selected candidate resource value according to the selected candidate resource value and the original consumed resources.

Further, the computer device can obtain the mapping function between the candidate resource value and the prediction adjustment coefficient, and substitute the selected candidate resource value and the original consumption resource into the mapping function, thereby calculating the prediction adjustment coefficient corresponding to the selected candidate resource value.

In this embodiment, determining the prediction adjustment coefficient corresponding to the candidate resource value according to the candidate resource value and the original consumption resource includes: determining the ratio of the candidate resource value to the original consumption resource as the prediction adjustment coefficient corresponding to the candidate resource value.

The computer device may determine the ratio of the selected candidate resource value to the original consumed resource, and further, the computer device may divide the selected candidate resource value by the original consumed resource to obtain the ratio. The computer device takes the ratio of the candidate resource value to the original consumed resource as the prediction adjustment coefficient corresponding to the candidate resource value.

Further, the computer device may obtain a mapping function between the candidate resource value and the prediction adjustment coefficient, and substitute the selected candidate resource value and the original consumed resource into the mapping function to calculate the ratio. The mapping function between candidate resource values and prediction adjustment coefficients is as follows:

λ ₂ = Candidate resource value/target_CPA (1)

Among them, λ ₂ is the prediction adjustment coefficient, and target_CPA is the original consumption resource.

In this embodiment, by arbitrarily selecting the candidate resource value from the candidate set, and calculating the corresponding prediction adjustment coefficient according to the selected candidate resource value and the original consumed resources, the prediction adjustment uniquely corresponding to the selected candidate resource value can be accurately obtained. coefficient.

In one embodiment, as shown in FIG. 3 , based on the predicted adjustment coefficient and the accumulated consumption resources, determining the adjustment expected accumulated consumption resources and expected conversion amount of the target information within the second preset time period, including:

Step 302 , based on the accumulated consumption resources, determine the corresponding adjustment expected accumulated consumption resources within the second preset time period when the target information is adjusted.

Adjusting the expected cumulative consumption resource refers to the expected cumulative consumption resource within a preset time period when the candidate resource value of the target information is adjusted, for example, adjusting the expected cumulative consumption resource corresponding to the second preset time period.

In this embodiment, the computer device acquires the resources actually consumed by the target information within the first preset time period, that is, the accumulated consumed resources. Further, the accumulated consumed resources refer to the accumulated consumed resources that are pushed within the first preset time period when the value of the candidate resource of the target information is adjusted.

The computer device obtains the historical adjustment coefficient of the target information within the first preset time period, and determines, based on the accumulated consumed resources and the historical adjustment coefficient, the corresponding unadjusted expected accumulated consumption resource within the first preset time period when the target information is not adjusted. Based on the first preset duration and the unadjusted expected accumulated consumption resource corresponding to the first preset duration, the unadjusted expected accumulated consumption resource corresponding to the second preset duration when the target information is not adjusted is determined. Based on the predicted adjustment coefficient and the unadjusted expected accumulated consumption resource corresponding to the first preset duration, the adjusted expected accumulated consumption resource corresponding to the second preset duration is determined when the target information is adjusted. The historical adjustment coefficient refers to a coefficient used to adjust the value of the candidate resource of the target information within the first preset time period of the target information.

Step 304: When the target information is not adjusted, the corresponding unadjusted expected consumption resources within the first preset time period are obtained.

The unadjusted expected consumption resource refers to the expected accumulated consumption resource within a preset time period under the condition that the value of the candidate resource of the target information is not adjusted. For example, the corresponding unadjusted expected accumulated consumption resources within the first preset time period and the corresponding unadjusted expected accumulated consumption resources within the second preset period of time.

In this embodiment, the value of the candidate resource of the target information is not adjusted means that the value of the candidate resource is fixed at 1. Adjusting the value of the candidate resource of the target information refers to adjusting the value of the candidate resource, for example, from 1 to 1.2, and from 1.2 to 0.8, but not limited thereto.

In this embodiment, the computer device acquires the historical adjustment coefficient of the target information within the first preset time period, and the corresponding consumed resources. According to the consumed resources and the historical adjustment coefficient, the unadjusted expected consumption resources of the target information within the first preset time period are determined.

Step 306 , based on the predicted adjustment coefficient and the unadjusted expected consumption resource, determine the adjustment expected consumption resource corresponding to the second preset time period when the target information is adjusted.

The ratio of the adjusted expected consumption resource corresponding to the second preset duration and the unadjusted expected consumption resource corresponding to the first preset duration is equal to the predicted adjustment coefficient. According to the relationship between the predicted adjustment coefficient, the unadjusted expected consumption resource and the adjusted expected consumption resource, and based on the known predicted adjustment coefficient and the unadjusted expected consumption resource, the computer device can directly calculate the target information when the second preset is adjusted. The adjustment corresponding to the duration is expected to consume resources.

Further, the computer device can calculate the product of the predicted adjustment coefficient and the unadjusted expected consumption resource, and use the product of the predicted adjustment coefficient and the unadjusted expected consumption resource to obtain the adjusted expected consumption corresponding to the second preset time period when the target information is adjusted. resource.

Step 308: Determine the expected conversion amount of the target information within the second preset time period according to the adjusted expected cumulative consumption resources and the adjusted expected consumption resources.

The computer device may calculate a ratio of the adjusted expected cumulative consumption resource to the adjusted expected consumption resource, and determine the ratio of the adjusted expected cumulative consumption resource to the adjusted expected consumption resource as the expected conversion amount.

In this embodiment, based on the accumulated consumption resources within the first preset time period under the adjustment condition, the expected accumulated consumption resources for adjustment within the second preset time period under the adjustment condition are calculated, which can be calculated according to the adjustment conditions under the past time period. The accumulated consumption resources are calculated, and the adjustment expected accumulated consumption resources in the case of adjustment in the future time period are calculated. According to the unadjusted expected consumption resource in the unadjusted case in the past time period and the predicted adjustment coefficient in the future time period, the adjustment expected consumption resource in the future time period under the adjustment condition is accurately calculated. Further, according to the adjustment expected cumulative consumption resources of the future time period and the adjustment expected consumption resources of the future time period, the expected conversion amount of the target information in the future time period can be accurately calculated.

In one embodiment, based on the predicted adjustment coefficient and the unadjusted expected consumption resource, determining the adjusted expected consumption resource corresponding to the second preset time period when the target information is adjusted, including: combining the predicted adjustment coefficient and the unadjusted expected consumption resource the product, which is determined to be the adjustment expected resource consumption corresponding to the second preset time period when the target information is adjusted;

Determining the expected conversion amount of the target information within the second preset time period according to the adjusted expected cumulative consumption resources and the adjusted expected consumption resources includes: determining the ratio of the adjusted expected cumulative consumption resources to the adjusted expected consumption resources as the target information in the second preset time period. Desired amount of conversions for a preset time period.

In this embodiment, the ratio of the adjusted expected consumption resources corresponding to the second preset duration to the unadjusted expected consumption resources corresponding to the first preset duration is equal to the predicted adjustment coefficient, as follows:

Wherein, CPA _E is the adjustment expected consumption resource corresponding to the target information within the second preset time period. CPA _1.0 is the unadjusted expected consumption resource of the target information within the first preset time period.

In this embodiment, the computer device divides the calculated and adjusted expected cumulative consumption resources by the adjusted expected consumption resources to obtain a ratio, and uses the ratio as the expected conversion amount. For example the following formula:

Wherein, _CVE is the expected conversion amount of the target information within the second preset time period. The cost _E is the expected cumulative consumption of resources for the adjustment of the target information within the second preset time period. CPA _E is the expected resource consumption for adjustment of the target information within the second preset time period.

In this embodiment, the product of the predicted adjustment coefficient and the unadjusted expected consumption resource is calculated to obtain the adjusted expected consumption resource, and the ratio of the expected cumulative consumption resource to the adjusted expected consumption resource is further calculated to obtain the expected conversion amount. According to the preset calculation method and Knowing the data, the required data can be calculated accurately and quickly.

In one embodiment, as shown in FIG. 4 , based on the accumulated consumption resources, determining the corresponding adjustment expected accumulated consumption resources within the second preset time period when the target information is adjusted, including:

Step 402: Obtain the historical adjustment coefficient of the target information within the first preset time period, and determine the unadjusted expected accumulated consumption resource corresponding to the first preset time period when the target information is not adjusted based on the accumulated consumed resources and the historical adjustment coefficient.

The computer device may acquire a coefficient for adjusting the candidate resource value of the target information within the first preset time period of the target information, that is, a historical adjustment coefficient. The computer device acquires the accumulated consumed resources of the target information within the first preset time period, and determines the ratio between the accumulated consumed resources and the square of the historical adjustment coefficient. The ratio between the accumulated consumption resource and the square of the historical adjustment coefficient is determined as the unadjusted expected accumulated consumption resource corresponding to the first preset time period when the target information is not adjusted.

Further, the computer device divides the accumulated consumption resource by the square of the historical adjustment coefficient to obtain the unadjusted expected accumulated consumption resource between the target information during adjustment within the first preset time period. For example, the computer equipment can obtain the following formula, and substitute the accumulated consumed resources and the historical adjustment coefficient into the following formula:

Wherein, cost _λ1 is the accumulated resource consumption of the target information within the first preset duration, and cost _1.0 is the corresponding unadjusted expected accumulated consumption resource within the first preset duration when the target information is not adjusted. λ ₁ is the historical adjustment coefficient.

Step 404 , based on the first preset duration and the unadjusted expected accumulated consumption resource corresponding to the first preset duration, determine the unadjusted expected accumulated consumption resource corresponding to the second preset duration when the target information is not adjusted.

The computer device determines, according to the first preset duration and the unadjusted expected cumulative consumption resources corresponding to the first preset duration, the expected cumulative consumption resources of the target information within the target duration, where the target duration includes the first preset duration and the second preset duration duration. Based on the expected accumulated consumption resource within the target duration and the unadjusted expected accumulated consumption resource corresponding to the first preset duration, the unadjusted expected accumulated consumption resource corresponding to the second preset duration when the target information is not adjusted is determined.

Step 406 , based on the predicted adjustment coefficient and the unadjusted expected cumulative consumption resource corresponding to the first preset duration, determine the adjusted expected cumulative consumption resource corresponding to the second preset duration when the target information is adjusted.

The product of the square of the predicted adjustment coefficient and the unadjusted expected accumulated consumption resource corresponding to the first preset duration is determined as the adjusted expected accumulated consumption resource corresponding to the second preset duration when the target information is adjusted.

Further, the computer equipment can obtain the following formula, and substitute the predicted adjustment coefficient and the unadjusted expected cumulative consumption resource into the formula, and calculate the adjusted expected cumulative consumption resource:

Wherein, cost _{E_1.0} is the unadjusted expected accumulated consumption resource corresponding to the second preset time period when the target information is not adjusted. The cost _E is the expected cumulative consumption of resources corresponding to the adjustment within the second preset time period when the target information is adjusted. λ ₂ is the prediction adjustment coefficient.

In this embodiment, the calculation of the adjusted expected accumulated consumption resources of the target information in the future time period takes into account the historical adjustment coefficients, accumulated consumption resources, unadjusted expected accumulated consumption resources in the past time period, and future time periods. Factors such as the predicted adjustment coefficient and the unadjusted expected cumulative resource consumption, fully consider the impact of various factors, making the prediction of the adjustment expected cumulative resource consumption in the future time period more accurate.

In one embodiment, based on the first preset duration and the unadjusted expected accumulated consumption resource corresponding to the first preset duration, determining the unadjusted expected accumulated consumption resource corresponding to the second preset duration when the target information is not adjusted, include:

According to the first preset duration and the unadjusted expected accumulated consumption resources corresponding to the first preset duration, determine the expected accumulated consumption resources of the target information within the target duration; the target duration includes the first preset duration and the second preset duration; Based on the expected accumulated consumption resource within the target duration and the unadjusted expected accumulated consumption resource corresponding to the first preset duration, the unadjusted expected accumulated consumption resource corresponding to the second preset duration when the target information is not adjusted is determined.

The computer device calculates the square of the first preset duration, obtains a weight corresponding to the square of the first preset duration, and obtains the weight of the first preset duration. The computer device calculates the product of the square of the first preset duration and the corresponding weight, calculates the product of the first preset duration and the corresponding weight, and determines the product of the square of the first preset duration and the corresponding weight, which is the same as the first preset duration. The sum of the products of the preset duration and the corresponding weight.

The computer device determines the ratio of the unadjusted expected cumulative consumption resource corresponding to the first preset duration to the sum of the two products as the expected cumulative consumption resource of the target information within the target duration. Further, the computer device divides the unadjusted expected cumulative consumption resource corresponding to the first preset duration by the sum of the two products to obtain a ratio.

Further, the computer device can obtain the following formula, and substitute the first preset duration and the unadjusted expected cumulative consumption resources into the formula, and calculate the expected cumulative consumption resources:

Wherein, cost _total is the expected cumulative resource consumption of the target information within the target duration, cost _1.0 is the corresponding unadjusted expected cumulative resource consumption within the first preset duration when the target information is not adjusted, and _t1 is the first preset duration.

In this embodiment, the computer device determines the difference between the expected accumulated consumption resource within the target duration and the unadjusted expected accumulated consumption resource corresponding to the first preset duration as the difference corresponding to the second preset duration when the target information is not adjusted The unadjusted expectations of the cumulative consumption of resources.

Further, the computer device can obtain the following formula, and substitute the expected cumulative consumption resources within the target duration and the unadjusted expected cumulative consumption resources corresponding to the first preset duration into the formula, and accurately calculate the unadjusted expected cumulative consumption resources:

cost _{E_1.0} = cost _total - cost _1.0 (7)

Wherein, cost _{E_1.0} is the unadjusted expected accumulated consumption resource corresponding to the second preset time period when the target information is not adjusted.

In one embodiment, acquiring the historical adjustment coefficient of the target information within the first preset time period includes:

Obtain the historical resource value used for pushing the target information within the first preset time period; and determine the historical adjustment coefficient of the target information within the first preset time period according to the historical resource value and the original consumption resource.

The historical resource value refers to the resource value consumed each time the target information obtains a push opportunity when the target information is pushed during the first preset duration. For example, if the target information has been pushed at 0.5 yuan, 1 yuan, and 3 yuan in the past 12 hours, the 0.5 yuan, 1 yuan, and 3 yuan are the historical resource values of the target information.

The historical adjustment coefficient refers to a coefficient used to adjust the value of the candidate resource of the target information within the first preset time period of the target information.

In this embodiment, the computer device can obtain the original resource consumption set by the information owner for the target information, and obtain the historical resource values when the candidate resource values of the target information are adjusted within the first preset time period. The computer device calculates the historical adjustment coefficient of the target information within the first preset time period according to the value of each historical resource and the original consumed resource.

For example, the computer equipment can sum the values of all historical resources and take the average value to obtain the average value of historical resources, and calculate the historical adjustment coefficient corresponding to the average value of historical resources according to the mapping function between the average value of historical resources, the original consumed resources and the historical adjustment coefficient .

In this embodiment, the corresponding historical adjustment coefficient is calculated based on the original consumption information of the target information and the historical resource value used to push the target information in the past time period, so that the adjustment coefficient used for adjustment in the past time period can be accurately calculated. The historical adjustment coefficient of the consumption resources of the target information.

In one embodiment, acquiring historical resource values used for pushing target information within a first preset time period includes: acquiring each historical resource value used for pushing target information within the first preset time period;

Determine the historical adjustment coefficient of the target information within the first preset time period according to the historical resource value and the original consumption resource, including: weighted average value of each historical resource value to obtain the historical resource average value; The ratio is determined as the historical adjustment coefficient of the target information within the first preset time period.

The historical resource average value refers to the average value of the historical resource values when the candidate resource value of the target information is adjusted within the first preset time period.

The computer device can obtain the original consumption resources set by the information owner for the target information, and obtain the historical resource values when the candidate resource values of the target information are adjusted within the first preset time period. The computer equipment obtains the weights corresponding to the values of each historical resource, and performs a weighted mean value process on the values of each historical resource to obtain the average value of the historical resources. Further, the computer equipment multiplies and sums each historical resource value and the corresponding weight, and divides the value obtained by the summation by the number of historical resource values to obtain the average value of historical resources.

The computer device may determine the ratio of the average value of historical resources to the original consumed resources, and determine the ratio of the average value of historical resources to the original consumed resources as the historical adjustment coefficient of the target information within the first preset time period. Further, the computer device may divide the mean value of historical resources by the original consumed resources to obtain a ratio, and use the ratio as a historical adjustment coefficient.

For example, the computer device can obtain the following formula, and calculate the historical adjustment coefficient according to the formula:

λ ₁ = historical resource average/target_CPA (8)

Among them, λ ₁ is the historical adjustment coefficient, and target_CPA is the original consumption resource.

In this embodiment, the value of each historical resource is weighted and averaged, and the ratio of the obtained average value of historical resources to the original consumed resources is used to determine the historical adjustment coefficient, so that the historical adjustment coefficient can be accurately determined according to the correlation between the average historical resource and the original consumed resources Determine the historical adjustment factor over the past time period.

In one embodiment, when the acquired target information is not adjusted, the corresponding unadjusted expected consumption resources within the first preset duration include:

Obtain the historical adjustment coefficient of the target information within the first preset time period and the corresponding consumed resources; according to the consumed resources and the historical adjustment coefficient, determine the unadjusted expected consumption resources of the target information within the first preset time period.

The consumed resource is the average cost consumed by one conversion within the first preset time period when the value of the candidate resource of the target information is not adjusted.

When the computer device obtains and adjusts the value of the candidate resource of the target information, it obtains the average cost consumed by one conversion within the first preset time period. The computer device acquires each historical resource value when the candidate resource value of the target information is adjusted within the first preset time period. According to the historical resource value and the original consumed resource, the historical adjustment coefficient of the target information within the first preset time period is determined.

In this embodiment, determining the historical adjustment coefficient of the target information within the first preset time period according to the historical resource value and the original consumption resource includes: performing a weighted average value on each historical resource value to obtain the historical resource average value; The ratio of the resource to the original consumption resource is determined as the historical adjustment coefficient of the target information within the first preset time period.

For example, the computer device can obtain the following formula, and calculate the historical adjustment coefficient according to the formula:

λ ₁ = historical resource average/target_CPA (8)

Among them, λ ₁ is the historical adjustment coefficient, and target_CPA is the original consumption resource.

Next, the computer device may calculate the ratio between the consumed resource and the historical adjustment coefficient, and determine the ratio as the unadjusted expected consumption resource of the target information within the first preset time period.

For example, the computer device can obtain the following formula, and calculate the unadjusted expected consumption resource of the target information within the first preset time period according to the formula:

为已消耗资源，CPA_1.0为目标信息在第一预设时长内的未调节期望消耗资源。in,

is the consumed resource, and CPA _1.0 is the unadjusted expected consumed resource of the target information within the first preset time period.

In this embodiment, by adjusting the value of the candidate resource of the target information, the historical adjustment coefficient of the target information in the first preset duration, and the average cost of obtaining one conversion, it is possible to accurately calculate the value of the candidate resource for the target information. In the case where the resource value is not adjusted, the target information obtains the average cost of one conversion in the first preset duration, so that the conversion cost in the case of adjustment can be accurately predicted by the conversion cost in the case of adjustment.

In one embodiment, based on the predicted adjustment coefficient and the unadjusted expected consumption resources, determining the adjustment expected consumption resources corresponding to the second preset time period when the target information is adjusted, including:

Obtain the historical adjustment coefficient of the target information within the first preset time period, and determine the unadjusted expected accumulated consumption resource corresponding to the target information within the first preset time period when the target information is not adjusted based on the accumulated consumed resources and the historical adjustment coefficient; obtain the unadjusted The smoothing factor corresponding to the expected consumption resource; according to the original consumption resource, the unadjusted expected consumption resource, the smoothing factor, the predicted adjustment coefficient, and the unadjusted expected cumulative consumption resource corresponding to the first preset duration, determine the target information adjustment when the second preset The adjustment corresponding to the duration is expected to consume resources.

The computer device may acquire a coefficient for adjusting the candidate resource value of the target information within the first preset time period of the target information, that is, a historical adjustment coefficient. The computer device acquires the accumulated consumed resources of the target information within the first preset time period, and determines the ratio between the accumulated consumed resources and the square of the historical adjustment coefficient. The ratio between the accumulated consumption resource and the square of the historical adjustment coefficient is determined as the unadjusted expected accumulated consumption resource corresponding to the first preset time period when the target information is not adjusted.

The computer device obtains the historical conversion amount corresponding to the target information in the historical duration adjacent to the target duration, and obtains the smoothing factor corresponding to the unadjusted expected consumption resource based on the target duration, the first preset duration and the historical conversion amount.

The computer device calculates the adjustment corresponding to the second preset time period when the target information is adjusted according to the original consumption resources, the unadjusted expected consumption resources, the smoothing factor, the predicted adjustment coefficient and the unadjusted expected accumulated consumption resources corresponding to the first preset time period. Expect to consume resources.

In this embodiment, according to the historical adjustment coefficient and the accumulated consumption resources in the past time period, the expected accumulated consumption of the unadjusted price in the past time period is accurately calculated under the assumption that the target information is not adjusted. The unadjusted expected accumulated consumption error in the past time period is relatively large, and the error can be reduced by the smoothing factor. Based on various factors such as original consumption resources, unadjusted expected consumption resources, smoothing factors, predicted adjustment coefficients, and unadjusted expected cumulative consumption resources, etc., estimate the adjustment expected consumption resources of target information in the future time period, and consider the data More comprehensive and more accurate estimates.

In one embodiment, according to the original consumption resources, the unadjusted expected consumption resources, the smoothing factor, the predicted adjustment coefficient, and the unadjusted expected accumulated consumption resources corresponding to the first preset duration, it is determined that the target information is adjusted within the second preset duration. The corresponding adjustment expectations consume resources, including:

Determine the predicted adjustment coefficient, the first product of the unadjusted expected cumulative consumption resource corresponding to the first preset duration and the unadjusted expected consumption resource, and the predicted adjustment coefficient, the smoothing factor, the unadjusted expected consumption resource and the second product of the original consumption resource ; Determine the sum of the unadjusted expected cumulative consumption resource corresponding to the third product of the smoothing factor and the unadjusted expected consumption resource and the first preset duration; multiply the first product by the second product, and divide by the third product and the first The sum of the unadjusted expected cumulative consumption resources corresponding to the preset time period is obtained, and the adjusted expected consumption resources corresponding to the second preset time period when the target information is adjusted is obtained.

The computer device multiplies the predicted adjustment coefficient, the unadjusted expected cumulative consumption resource corresponding to the first preset duration, and the unadjusted expected consumption resource to obtain a first product. The computer device multiplies the predicted adjustment coefficient, the smoothing factor, the unadjusted expected consumption resource and the original consumption resource to obtain a second product.

The computer device multiplies the smoothing factor and the unadjusted expected consumption resource corresponding to the first preset duration to obtain a third product. The computer device calculates the third product and the sum of the unadjusted expected accumulated consumption resources corresponding to the first preset duration.

The computer device multiplies the first product and the second product, divides the product obtained by the multiplication by the third product and the sum of the unadjusted expected cumulative consumption resources corresponding to the first preset duration, and obtains the target information when the second preset is adjusted. Set the corresponding adjustment expectations within the duration to consume resources.

For example, the computer device can obtain the following formula, and substitute the original consumption resources, the unadjusted expected consumption resources corresponding to the first preset duration, the smoothing factor, the predicted adjustment coefficient, and the unadjusted expected cumulative consumption resources corresponding to the first preset duration into the formula , to obtain the adjustment expected resource consumption corresponding to the second preset time period when the target information is adjusted:

Wherein, CPA _E is the adjustment expected consumption resource corresponding to the second preset time period. λ ₂ is the predicted adjustment coefficient, target_CPA is the original consumption resource, CPA _1.0 is the unadjusted expected consumption resource corresponding to the first preset duration, w is the smoothing factor, and cost _1.0 is the unadjusted expected cumulative consumption resource corresponding to the first preset duration. .

In this embodiment, according to the known target information in the past time period, unadjusted expected consumption resources, unadjusted expected accumulated consumption resources, smoothing factor, and the original consumption resources set by the information owner, it can be accurately predicted that in the future time period The adjustment of target information is expected to consume resources. According to the known information in the past time period, the resource consumption of the target information in the future time period is estimated, so as to fully plan the push of the target information in the future time period, so as to realize the accurate push of the target information.

In one embodiment, the smoothing factor is obtained through a smoothing factor determining step, and the smoothing factor determining step includes: acquiring the historical conversion amount corresponding to the target information in the historical duration adjacent to the target duration; A preset duration and historical conversion amount are obtained, and a smoothing factor corresponding to the unadjusted expected resource consumption is obtained.

The historical duration is the duration that is adjacent to the target duration and has the same duration. For example, if the target duration is 00:00-24:00 on December 2, 2020, the historical duration is 00:00-24:00 on December 1, 2020. Historical conversions refers to the number of conversions or conversions for a target message over a historical period of time.

In this embodiment, the computer device may determine the historical duration that is adjacent to the target duration and has the same duration. The computer device may obtain the historical conversion amount of the target information within the historical period. The computer device may determine a ratio of the first preset duration to the target duration, and calculate a cubic sum of the difference between 1 and the ratio. The computer device uses the product of the cubic sum and the historical conversion amount as a smoothing factor corresponding to the unadjusted expected consumption resource.

Further, the computer device may obtain the smoothing factor function, and substitute the target duration, the first preset duration and the historical conversion amount into the smoothing factor function to obtain the smoothing factor. The smoothing factor function is given by the following formula:

Wherein, w is a smoothing factor, t ₁ is the first preset duration, t _total is the target duration, and CV′ is the historical conversion amount of the target information in the historical duration. The computer device substitutes the target duration, the first preset duration and the historical conversion amount into the above formula to obtain the smoothing factor w.

In this embodiment, by calculating a smoothing factor, the adjustment expected resource consumption of the target information within the first preset time period is smoothed by the smoothing factor, so as to smooth the estimated consumption resource of the target information in the past time period, It can reduce the error of estimated resource consumption.

In one embodiment, according to the converted amount, the accumulated consumption resources, the expected conversion amount and the adjusted expected accumulated consumption resources, determining the target consumption resources of the target information within the target duration, including:

Determine the sum of the accumulated consumption resources and the adjusted expected accumulated consumption resources, as well as the sum of the converted amount and the expected conversion amount; The ratio is determined as the target resource consumption of the target information within the target duration.

The computer device calculates the sum of the accumulated consumption resources and the adjusted expected accumulated consumption resources, and calculates the sum of the converted amount and the expected conversion amount. The computer device divides the two results to obtain a ratio, and uses the ratio as the target resource consumption of the target information within the target duration.

Further, the following formula can be obtained, and the converted amount, the accumulated consumption resource, the expected conversion amount and the adjusted expected cumulative consumption resource are substituted into the formula to obtain the target consumption resource:

为目标信息在第一预设时长内的已累积消耗资源，

为目标信息在第一预设时长内的已累积转化量。Among them, E _{(CPA_final)} is the target consumption resource corresponding to the target information within the target duration. The cost _E is the expected cumulative consumption of resources for the adjustment of the target information within the second preset time period. CV _E is the expected conversion amount of the target information within the second preset time period.

is the accumulated consumption resource of the target information within the first preset time period,

is the accumulated conversion amount of the target information within the first preset time period.

In this embodiment, by using the relationship between the converted amount, the accumulated consumption resources, the expected conversion amount, and the adjusted expected cumulative consumption resources, it is predicted that the target information can be obtained within the target duration when the candidate resource value selected this time is used. Transform the resource expected to be consumed at one time, so as to determine whether the resource expected to be consumed is eligible to perform the corresponding operation. If the condition is met, it means that accurate push of information can be achieved based on the value of the candidate resource selected in this round. If the condition is not met, it means that the value of the candidate resource selected in this round cannot achieve accurate push of information, and the next round of selection of the candidate resource value will be performed. and calculate.

In one embodiment, the method further includes:

When the target consumption resource and the original consumption resource do not meet the information push conditions, select another candidate resource value from the candidate set, and return to the step of determining the prediction adjustment coefficient corresponding to the selected candidate resource value, so as to obtain the target information in the target duration When the target consumption resource and the original consumption resource meet the information push conditions, the target information is pushed according to the selected candidate resource value.

When the computer device determines that the target consumption resource and the original consumption resource do not meet the information push conditions, it indicates that the selected candidate resource value is not suitable for pushing target information within the second preset time period. The computer device can arbitrarily select another candidate resource value from the candidate set, and return and execute to determine the prediction adjustment coefficient corresponding to the selected candidate resource value, until the calculated target consumption resources and the original consumption resources meet the information push conditions, according to the selected The candidate resource value pushes the target information.

When the target consumption resource is calculated based on the selected candidate resource value, and the target consumption resource and the original consumption resource still do not meet the information push conditions, the candidate resource value in the candidate set is traversed. The traversal is stopped until it is calculated based on the selected candidate resource value that the target consumption resource is the same as the original consumption resource, and the target information is pushed within the second preset time period according to the selected candidate resource value.

In one embodiment, when the target consumption resource is not equal to the original consumption resource, it is determined that the target consumption resource and the original consumption resource do not meet the information push condition.

In other embodiments, when the difference between the target consumption resource and the original consumption resource is greater than a threshold, it is determined that the target consumption resource and the original consumption resource do not meet the information push condition.

In this embodiment, the candidate resource values are selected by traversal from the candidate set, and the target consumption resources consumed by one conversion within the target duration are calculated based on the selected candidate resource values. The value of the candidate resource, so as to realize the accurate push of information according to the value of the candidate resource.

In one embodiment, the method further includes: determining the respective expected revenue of each target information when the information is pushed according to the respective candidate resource values; and pushing the target information corresponding to the expected revenue satisfying the revenue condition.

Expected revenue refers to the revenue obtained when the target information is pushed according to the value of the candidate resource.

When there are multiple target information, for each target information and the corresponding candidate resource value, the computer device calculates the expected benefit obtained when the target information is pushed according to the corresponding candidate resource value. The computer device can obtain the income conditions, judge whether each expected income meets the expected conditions, and push the target information corresponding to the expected incomes that meet the expected conditions according to the corresponding candidate resource values.

Further, the computer device calculates the corresponding expected revenue of each target information when the information is pushed within the second preset time period according to the respective candidate resource values. The computer device pushes the target information corresponding to the expected revenue that satisfies the revenue condition within the second preset time period.

For example, the computer equipment can calculate the expected benefit of the target information according to the benefit calculation formula:

ecpm=Target_CPA*pCTR*pCVR*1000*candidate resource value (13)

Among them, ecpm (effective cost per mille) is the expected income obtained by pushing the target information according to the value of the candidate resource, for example, the income obtained by 1000 pushes. pCTR (predict click through rate, estimated click through rate) refers to the estimated probability of being clicked after the target information is pushed. This estimated CTR can be predicted by a CTR model. CVR (Conversion Rate, conversion rate) refers to the proportion of users who click and convert the target information, that is, the number of conversions of the target information divided by the number of clicks on the target information. pCVR (predict conversion rate, estimated conversion rate) refers to the estimated conversion rate after the target information is pushed, which is predicted by the conversion rate model.

In this embodiment, the expected income of the target information is used as a condition for information push, and the target information corresponding to the expected income satisfying the income condition is pushed, so as to achieve accurate information push.

In one embodiment, the target information is a target advertisement, the original consumption resource is the original conversion unit price, the candidate resource value is the candidate bid price, the accumulated consumption resource is the converted total price, and the adjusted expected accumulated consumption resource is the price adjustment expected conversion total price, The target consumption resource is the target conversion unit price;

When the target consumption resource and the original consumption resource meet the information push conditions, the target information is pushed according to the value of the candidate resource, including: when the target conversion unit price is the same as the original conversion unit price, determining the advertisement placement position corresponding to the candidate bid; within the ad placements to serve targeted ads.

An ad placement location refers to the location where an ad is placed, including any location where an ad can be placed. For example, the location for advertising in the information flow, the display screen for advertising in subway stations, bus stop signs, or the display screen for advertising on buildings, the terminal interface, the location for advertising in applications, etc., but Not limited to this.

In this embodiment, the computer device obtains the original conversion unit price and the candidate bid price corresponding to the target advertisement, and determines the prediction adjustment coefficient corresponding to the candidate bid price. The computer device obtains the converted volume and the converted total price of the target advertisement within the first preset time period, and determines the expected total conversion price and expected conversion price of the target advertisement within the second preset time period based on the predicted adjustment coefficient and the converted total price conversion volume.

The computer device determines the target resource consumption of the target advertisement within the target duration according to the converted volume, the converted total price, the expected conversion volume and the price adjusted expected conversion total price; the target duration includes the second preset duration and the first preset duration. When the target conversion unit price is the same as the original conversion unit price, the computer device determines the advertisement placement position corresponding to the candidate bid, and places the target advertisement at the advertisement placement position within the second preset time period.

In this embodiment, the converted amount and cost of the target advertisement by users in the past time period are fully considered, and according to the converted amount, cost, historical price adjustment coefficient and other factors in the past time period, it is accurately estimated that the user's interest in the target advertisement in the future time period is accurately estimated. The estimated conversion volume and cost of the targeted ad. And based on the user's conversion volume and cost of the advertisement in the past time period, as well as the user's estimated conversion volume and cost of the target advertisement in the future time period, more accurately predict the target advertisement's conversion cost in the future time period. , so as to more accurately determine the price of target advertising. At the same time, if the target conversion unit price and the original conversion unit price satisfy the advertisement placement conditions, it means that the predicted cost of conversion of the target advertisement in the future time period is roughly the same as the cost defined by the advertiser, and the corresponding advertisement is placed according to the candidate bid. Positioning targeted advertisements can achieve accurate advertisement placement, thereby obtaining higher advertisement conversions and improving the efficiency of advertisement placement under the condition of meeting the cost provided by the advertisement owner.

As shown in FIG. 5 , it is a schematic diagram of an interface for placing a target advertisement in an advertisement placement position in an information flow page in one embodiment. The second preset duration is 12:00-24:00, and the computer device places advertisements corresponding to the candidate bids in the advertisement placement positions 502 and 504 of the information flow page 500, such as advertisement 1 in the advertisement placement position 502, advertisement placement position 504 advertisement 2, and the delivery time is 12:00-24:00. When the user goes to the information flow page, the corresponding advertisement is displayed in the advertisement placement position. Friends in the user's friend list can also browse advertisements displayed on the user's information flow page through the information flow page.

As shown in FIG. 6 , it is a schematic diagram of an interface for placing a target advertisement on an advertisement display screen in another embodiment.

The scene is a plurality of advertisement display screens in the subway station, the second preset time period is 12:00-24:00, and the computer equipment places advertisements corresponding to candidate bids on the advertisement display screens 602, 604 and 606, such as advertisement display screens Advertisement A of 602, advertisement B of advertisement display screen 604, advertisement C of advertisement display screen 606, and the delivery time is 12:00-24:00. When the user goes to the information flow page, the corresponding advertisement is displayed in the advertisement placement position.

In one embodiment, the first preset duration is the duration from the start point of the target duration to the current moment, the second preset duration is the duration from the current moment to the end of the target duration, and the target duration is a periodic fixed duration.

In this embodiment, the information push method can be applied to advertisement placement, and it is possible to more accurately predict the amount of conversions and costs of advertisements by users in the past time period, and the estimated conversion amount and cost of advertisements by users in the future time period. The cost consumed by the conversion of the advertisement in the future time period, so as to more accurately determine the bidding price of the information push, and realize the accurate delivery of the advertisement based on the candidate bidding. At the same time, the target conversion unit price and the original conversion unit price meet the advertising push conditions, which means that the cost of converting the advertisement in the future time period is roughly the same as the cost defined by the advertiser, so that the cost can be obtained under the condition of satisfying the cost provided by the advertiser. High advertising conversion volume, improve the efficiency of advertising.

In one embodiment, the first preset duration is the duration from the starting point of the target duration to the current moment, the second preset duration is the duration from the current moment to the end of the target duration, and the target duration is a periodic fixed duration.

The starting point is the start moment of the target duration, and is also the start moment of the first preset duration. The end point refers to the end moment of the target duration, and is also the end moment of the second preset duration. The current moment is the end moment of the first preset duration, and is also the start moment of the second preset duration. The target duration is a periodic fixed duration.

In this embodiment, the target duration is a periodic fixed duration, and the information push method can be based on the actual situation of the target information within each target duration, such as the converted amount, accumulated consumption resources and other candidate resources for the target information The value is adjusted in time, so that the target information is pushed with the appropriate candidate resource value, and the accurate push of the target information is automatically realized.

As shown in FIG. 7 , it is a schematic diagram of the principle of an information push method in an embodiment.

After receiving the information request from the traffic end, the information delivery engine recalls the information directed to the traffic end from the information database. Through the solutions in the above embodiments, the target resource consumption of the recalled target information is determined, and when the target resource consumption of the target information is equal to the original consumption resource, the selected candidate resource value is used as the target resource value corresponding to the target information.

For each target information recalled, the information delivery engine performs rough selection and sorting on each target information. Rough selection sorting is used to filter the target information to achieve the effect of selecting hundreds of thousands of miles.

The steps of rough sorting are as follows:

The information delivery engine uses the rough selection model to predict the rough selection estimated click-through rate and the rough selection estimated conversion rate of each target information pushed according to the candidate resource value, so as to calculate the expected revenue of each target information in the rough selection sorting stage, that is,

ec1=Target_CPA*liteCTR*liteCVR*1000*candidate resource value (14)

Among them, ec1 is the expected revenue of the target information in the rough selection stage, liteCTR is the estimated click-through rate of the target information in the rough selection stage, and liteCVR is the estimated conversion rate of the target information in the rough selection stage. The model in the rough selection stage can use NPL (Natural Language Processing, natural language processing), recommendation algorithm, machine learning, distributed learning and other methods to predict the target information in the rough selection stage to estimate the click rate and the conversion rate of the rough selection.

According to the expected income of each target information, a preset number of target information is selected in the rough selection stage to enter the selection stage. For example, select the target information whose expected income is higher than the preset income to enter the selection stage, and perform selection sorting. The selection sorting is used to further filter the target information to achieve the effect of one in a hundred.

The steps for selection sorting are as follows:

The information delivery engine uses the selection model to predict the selected estimated click-through rate and the selected estimated conversion rate of each target information pushed according to the candidate resource value, so as to calculate the expected revenue of each target information in the selection and sorting stage, that is,

ec1=Target_CPA*pCTR*pCVR*1000*candidate resource value (15)

Among them, ec2 is the expected revenue of the target information in the selection stage, pCTR is the estimated click-through rate of the target information in the selection stage, and pCVR is the estimated conversion rate of the target information in the selection stage. The model in the selection stage can use NPL, recommendation algorithm, machine learning, distributed learning and other methods to predict the target information in the selection stage to select the estimated click-through rate and the selection estimated conversion rate.

The information delivery engine sorts the expected income of each target information in the selection stage, and delivers the target information that meets the income conditions to the traffic end.

In one embodiment, an information push method is provided, comprising:

In step (a1), the converted amount and the accumulated consumption resources of the target information within the first preset time period are obtained, and the historical resource value used for pushing the target information within the first preset time period is obtained.

In step (a2), the original consumption resource corresponding to the target information is obtained, and the historical adjustment coefficient of the target information within the first preset duration is determined according to the historical resource value and the original consumption resource.

In step (a3), based on the accumulated consumed resources and the historical adjustment coefficient, determine the corresponding unadjusted expected accumulated consumption resources within the first preset time period when the target information is not adjusted.

Step (a4), according to the first preset duration and the unadjusted expected accumulated consumption resource corresponding to the first preset duration, determine the expected accumulated consumption resource of the target information within the target duration, and the target duration includes the first preset duration and the first preset duration. 2. Preset duration.

Step (a5), based on the expected cumulative consumption resource in the target duration and the unregulated expected cumulative consumption resource corresponding to the first preset duration, determine the unregulated expected cumulative consumption corresponding to the second preset duration when the target information is not adjusted resource.

In step (a6), a candidate resource value for pushing target information is selected from the candidate set, and a prediction adjustment coefficient corresponding to the candidate resource value is determined according to the candidate resource value and the original consumed resource.

Step (a7), based on the predicted adjustment coefficient and the unadjusted expected accumulated consumption resource corresponding to the first preset duration, determine the adjusted expected accumulated consumption resource corresponding to the second preset duration when the target information is adjusted.

Step (a8): Acquire the consumed resources of the target information within the first preset time period, and determine the unadjusted expected consumption resources of the target information within the first preset time period according to the consumed resources and the historical adjustment coefficient.

In step (a9), based on the accumulated consumed resources and the historical adjustment coefficient, determine the corresponding unadjusted expected accumulated consumption resources within the first preset time period when the target information is not adjusted.

Step (a10): Obtain the historical conversion amount corresponding to the target information in the historical duration adjacent to the target duration.

In step (a11), based on the target duration, the first preset duration and the historical conversion amount, a smoothing factor corresponding to the unadjusted expected consumption resource is obtained.

Step (a12), according to the original consumption resources, the unadjusted expected consumption resources, the smoothing factor, the predicted adjustment coefficient and the unadjusted expected cumulative consumption resources corresponding to the first preset time length, determine the target information adjustment within the second preset time length. The corresponding adjustment expectations consume resources.

Step (a13), according to the expected cumulative consumption of resources and the adjustment of the expected consumption of resources, to determine the expected conversion amount of the target information within the second preset time period.

Step (a14): Determine the target consumption resource of the target information within the target duration according to the converted amount, the accumulated consumption resource, the expected conversion amount and the adjusted expected cumulative consumption resource.

Step (a15), when the target consumption resources and the original consumption resources meet the information push conditions, determine the expected income corresponding to each target information when the information is pushed according to the respective candidate resource values; the target corresponding to the expected income that satisfies the income conditions The information is pushed within the second preset time period according to the value of the candidate resource.

Step (a16), when the target consumption resource and the original consumption resource do not meet the information push conditions, select another candidate resource value from the candidate set, and return to the step of determining the prediction adjustment coefficient corresponding to the selected candidate resource value to obtain The target consumption resource of the target information within the target duration; when the target consumption resource and the original consumption resource meet the information push conditions, the target information is pushed within the second preset duration according to the selected candidate resource value.

In this embodiment, the influence of various factors is taken into consideration through the conversion amount and cost of information by users in the past time period, and the estimated conversion amount and cost of information by users in the future time period, so as to more accurately predict the information The cost of obtaining conversions in the future time period, so as to more accurately determine the price of information push, and achieve accurate push information. At the same time, the target consumption resources and the original consumption resources meet the information push conditions, indicating that the cost of obtaining and transforming the predicted information in the future time period is roughly the same as the cost defined by the information owner, thus satisfying the cost provided by the information owner. Under the condition of higher information conversion, improve the efficiency of information push.

The present application also provides an application scenario where the above-mentioned information push method is applied to advertisement placement. Specifically, the application of the information push method in the application scenario is shown in FIG. 8 , including:

The target information is the target advertisement, the original consumption resource is the original conversion unit price, the candidate resource value is the candidate bid, the accumulated consumption resource is the converted total price, the consumed resource is the converted unit price, and the adjusted expected cumulative consumption resource is the price adjustment expected conversion total The target consumption resource is the target conversion unit price. The predicted adjustment coefficient is the predicted price adjustment coefficient, the historical adjustment coefficient is the historical price adjustment coefficient, the first preset duration is the duration from 0 to the current time t, and the second preset duration is the duration from the current time t to 24 o'clock.

Collect advertising data, and obtain the original conversion unit price target_CPA defined by the advertiser to obtain 1 advertising conversion. At time 0-t, the target advertisement adjusts the candidate bid of the target advertisement through the historical price adjustment coefficient λ ₁ to place the target advertisement, that is, the price of the target advertisement is adjusted through the historical price adjustment coefficient λ ₁ . The advertising conversion volume accumulated by the target advertisement at time 0-t is the converted volume, and the accumulated cost is the total converted price

Obtain the candidate bids used to place the target advertisement in the time 0-t, and obtain the weighted average of each candidate bid. According to the weighted average divided by the original conversion unit price target_CPA, the historical price adjustment coefficient λ ₁ at time 0-t is obtained. which is:

λ ₁ = weighted average of each candidate bid/target_CPA

和实际的历史调价系数λ₁，确定目标广告在0-t时刻内的未调价期望转化总价cost_1.0。即：Assuming that the target advertisement does not adjust the candidate bid at time 0-t, that is, on the basis of placing the target advertisement according to the price adjustment coefficient of 1.0, according to the actual converted total price at time 0-t

and the actual historical price adjustment coefficient λ ₁ , to determine the unadjusted expected conversion price cost _1.0 of the target advertisement at time 0-t. which is:

According to the duration t ₁ at time 0-t, and the unadjusted expected total conversion price cost _1.0 at time 0-t, calculate the expected total conversion price cost _total of the target advertisement at 0-24, namely:

Based on the expected total conversion price cost _total at 0-24 points and the unadjusted expected total conversion price cost _1.0 at time 0-t, calculate the unadjusted expected total conversion price cost _{E_1} at point t-24 when the target advertisement has not been adjusted in price _.0 , that is:

cost _{E_1.0} = cost _total - cost _1.0

A candidate bid for pushing the target advertisement is selected from the candidate set, and a predicted price adjustment coefficient λ ₂ corresponding to the candidate bid is determined according to the candidate bid and the original conversion unit price target_CPA. which is:

λ ₂ = candidate bids selected in the candidate set/target_CPA

According to the predicted price adjustment coefficient λ ₂ and the unadjusted expected total conversion price cost _{E_1.0} within point t-24, calculate the expected total conversion price cost _E of target advertisement price adjustment within t-24 point under the condition of price adjustment, namely:

根据已转化单价

和历史调价系数λ₁，计算出在假设未调价情况下，目标广告在0-t时刻内的未调价期望转化单价CPA_1.0。即：Obtain the converted unit price at time 0-t when the price of the target advertisement is adjusted by the historical price adjustment coefficient λ ₁

According to the converted unit price

and the historical price adjustment coefficient λ ₁ to calculate the expected conversion unit price CPA _1.0 of the target advertisement in the 0-t time period without price adjustment under the assumption that the price has not been adjusted. which is:

Obtain the historical conversion volume CV' of the target advertisement in the adjacent previous 0-24 points from 0-24 points. According to the duration t _total of 0-24 points, the duration t ₁ at moment 0-t and the historical conversion volume CV', Calculate the smoothing factor w corresponding to the unadjusted expected conversion unit price CPA _1.0 at time 0-t, namely:

According to the original conversion unit price target_CPA, the unadjusted expected conversion unit price CPA _1.0 at time 0-t, the smoothing factor w, the predicted price adjustment coefficient λ ₂ and the unadjusted expected total conversion price cost _1.0 at time 0-t, determine when the target advertisement price is adjusted. The expected conversion unit price CPA _E of the corresponding price adjustment within t-24 point, namely:

According to the price adjustment expected conversion total price cost _E within t-24 point and the price adjustment expected conversion unit price CPA _E within t-24 point, determine the expected conversion volume CV _E of the target advertisement within t-24 point:

已转化总价

期望转化量CV_E和期望转化总价cost_E，确定目标广告在0-24点内的目标转化单价E_{(CPA_final)}，即：Based on conversions

Total price converted

The expected conversion volume CV _E and the expected conversion total cost _E , determine the target conversion unit price E _{(CPA_final)} of the target advertisement within 0-24 points, namely:

When E _{(CPA_final)} = target_CPA, traverse the candidate bids in the candidate set according to the above steps, stop traversing until the final target conversion unit price is the same as the original conversion unit price, and use the candidate bid when the target conversion unit price and the original conversion unit price are the same The bid is used as the target bid corresponding to the target advertisement, and the target bid is used to act online.

When there are multiple target advertisements, determine the estimated click-through rate pCTR and estimated conversion rate pCVR of the multiple target advertisements placed within t-24 using their respective target bids to calculate the expectations obtained by placing each target advertisement 1000 times. Yield ecpm, that is:

ecpm=Target_CPA*pCTR*pCVR*1000*target bid

The expected income corresponding to each target advertisement is sorted, and the target advertisement corresponding to the expected income higher than the preset income is placed within t-24 points according to the respective advertisement placement positions.

It should be understood that, although the steps in the flowcharts of FIGS. 2-4 and 8 are displayed in sequence according to the arrows, these steps are not necessarily executed in the sequence indicated by the arrows. Unless explicitly stated herein, the execution of these steps is not strictly limited to the order, and these steps may be performed in other orders. Moreover, at least a part of the steps in FIGS. 2-4 and 8 may include multiple steps or multiple stages. These steps or stages are not necessarily executed at the same time, but may be executed at different times. These steps or The order of execution of the stages is also not necessarily sequential, but may be performed alternately or alternately with other steps or at least a portion of the steps or stages in the other steps.

In one embodiment, as shown in FIG. 9 , an apparatus for pushing information is provided, and the apparatus may adopt a software module or a hardware module, or a combination of the two to become a part of a computer device, and the apparatus specifically includes: an acquisition module 902 , a first determination module 904, and a push module 908, wherein:

The obtaining module 902 is configured to obtain the original consumption resource and the candidate resource value corresponding to the target information, and determine the prediction adjustment coefficient corresponding to the candidate resource value.

The first determination module 904 is configured to obtain the converted amount and the accumulated consumption resources of the target information within the first preset time period, and determine the adjustment of the target information within the second preset time period based on the predicted adjustment coefficient and the accumulated consumption resources Expected cumulative consumption of resources and expected conversions.

The second determining module 906 is configured to determine the target consumption resource of the target information within the target duration according to the converted amount, the accumulated consumption resource, the expected conversion amount and the adjusted expected cumulative consumption resource; the target duration includes the second preset duration and The first preset duration.

The pushing module 908 is configured to push the target information according to the value of the candidate resource when the target consumption resource and the original consumption resource meet the information pushing condition.

In this embodiment, the adjustment coefficient corresponding to the candidate resource value is determined by acquiring the original consumption resource corresponding to the target information and the candidate resource value. According to the converted amount and accumulated consumption resources of the target information in the past time period, and the predicted adjustment coefficient, the expected conversion amount and the expected accumulated consumption resources of the target information in the future time period are accurately calculated. According to the conversion amount and accumulated consumption resources in the past time period, the expected conversion amount in the future and the expected cumulative consumption resources, the average consumption resources of the target information in the total time period including the past time period and the future time period can be further calculated, Thereby, it is determined whether the average consumption resource meets the conditions, so as to perform the corresponding operation. When the average consumption resources and the original consumption resources meet the information push conditions, it means that the value of the candidate resource is the most suitable resource value for pushing the target information, then the target information is pushed according to the value of the candidate resource, which can achieve accurate information push, and at the same time A higher amount of information conversion can be obtained at the same cost.

In the traditional way, the information is pushed according to the custom push method, and the conversion volume and conversion cost in different time periods are not considered. In this embodiment, the conversion volume and cost of information by users in the past time period are considered, as well as the estimated conversion volume and cost of information by users in the future time period. Based on the influence of various factors, more accurate predictions can be made. The cost of obtaining and converting information in the future time period, so as to more accurately determine the price of information push and realize accurate push of information. At the same time, the target consumption resources and the original consumption resources meet the information push conditions, indicating that the cost of obtaining and transforming the predicted information in the future time period is roughly the same as the cost defined by the information owner, thus satisfying the cost provided by the information owner. Under the condition of higher information conversion, improve the efficiency of information push.

In one embodiment, the obtaining module 902 is further configured to: obtain the original consumption resource corresponding to the target information, and select the candidate resource value for pushing the target information from the candidate set; determine the candidate resource value according to the candidate resource value and the original consumption resource Corresponding prediction adjustment coefficient.

In this embodiment, by arbitrarily selecting the candidate resource value from the candidate set, and calculating the corresponding prediction adjustment coefficient according to the selected candidate resource value and the original consumed resources, the prediction adjustment uniquely corresponding to the selected candidate resource value can be accurately obtained. coefficient.

In one embodiment, the obtaining module 902 is further configured to: determine the ratio of the candidate resource value to the original consumed resource as the prediction adjustment coefficient corresponding to the candidate resource value.

In one embodiment, the first determining module 904 is further configured to: based on the accumulated consumption resources, determine the corresponding adjustment expected accumulated consumption resources within the second preset time period when the target information is adjusted; when the acquired target information is not adjusted, in the first The unadjusted expected consumption resource corresponding to a preset time period; based on the predicted adjustment coefficient and the unadjusted expected consumption resource, determine the adjusted expected consumption resource corresponding to the target information adjustment within the second preset time period; according to the adjustment expected cumulative consumption Resource and adjust the expected consumption of resources, and determine the expected conversion amount of the target information within the second preset time period.

In this embodiment, based on the accumulated consumption resources within the first preset time period under the adjustment condition, the expected accumulated consumption resources for adjustment within the second preset time period under the adjustment condition are calculated, which can be calculated according to the adjustment conditions under the past time period. The accumulated consumption resources are calculated, and the adjustment expected accumulated consumption resources in the case of adjustment in the future time period are calculated. According to the unadjusted expected consumption resource in the unadjusted case in the past time period and the predicted adjustment coefficient in the future time period, the adjustment expected consumption resource in the future time period under the adjustment condition is accurately calculated. Further, according to the adjustment expected cumulative consumption resources of the future time period and the adjustment expected consumption resources of the future time period, the expected conversion amount of the target information in the future time period can be accurately calculated.

In one embodiment, the first determining module 904 is further configured to: determine the product of the predicted adjustment coefficient and the unadjusted expected consumption resource as the adjustment expected consumption resource corresponding to the second preset time period when the target information is adjusted;

The second determining module 906 is further configured to: determine the ratio of the adjusted expected accumulated consumption resource to the adjusted expected consumption resource as the expected conversion amount of the target information within the second preset time period.

In this embodiment, the product of the predicted adjustment coefficient and the unadjusted expected consumption resource is calculated to obtain the adjusted expected consumption resource, and the ratio of the expected cumulative consumption resource to the adjusted expected consumption resource is further calculated to obtain the expected conversion amount. According to the preset calculation method and Knowing the data, the required data can be calculated accurately and quickly.

In one embodiment, the first determining module 904 is further configured to: obtain a historical adjustment coefficient of the target information within a first preset time period, and determine, based on the accumulated consumed resources and the historical adjustment coefficient, to determine the first preset time period when the target information is not adjusted. Set the unadjusted expected cumulative consumption resources corresponding to the time length; based on the first preset time length and the unadjusted expected cumulative consumption resources corresponding to the first preset time length, determine the unadjusted expected cumulative consumption resources corresponding to the second preset time length when the target information is not adjusted. Adjusting the expected cumulative consumption resource; based on the predicted adjustment coefficient and the unadjusted expected cumulative consumption resource corresponding to the first preset duration, determining the adjusted expected cumulative consumption resource corresponding to the second preset duration when the target information is adjusted.

In this embodiment, the calculation of the adjusted expected accumulated consumption resources of the target information in the future time period takes into account the historical adjustment coefficients, accumulated consumption resources, unadjusted expected accumulated consumption resources in the past time period, and future time periods. Factors such as the predicted adjustment coefficient and the unadjusted expected cumulative resource consumption, fully consider the impact of various factors, making the prediction of the adjustment expected cumulative resource consumption in the future time period more accurate.

In one embodiment, the first determining module 904 is further configured to: determine the expected cumulative consumption resource of the target information within the target duration according to the first preset duration and the unadjusted expected cumulative consumption resource corresponding to the first preset duration; the target The duration includes a first preset duration and a second preset duration; based on the expected accumulated consumption resources within the target duration and the unadjusted expected accumulated consumption resources corresponding to the first preset duration, it is determined that when the target information is not adjusted, the second preset duration is The corresponding unadjusted expected cumulative consumption resource within the

In one embodiment, the first determining module 904 is further configured to: acquire the historical resource value used for pushing the target information within the first preset time period; The historical adjustment factor for the duration.

In this embodiment, the corresponding historical adjustment coefficient is calculated based on the original consumption information of the target information and the historical resource value used to push the target information in the past time period, so that the adjustment coefficient used for adjustment in the past time period can be accurately calculated. The historical adjustment coefficient of the consumption resources of the target information.

In one embodiment, the first determining module 904 is further configured to: obtain the historical resource values used for pushing the target information within the first preset time period; perform a weighted average of the historical resource values to obtain the historical resource average value; The ratio of the average value of historical resources to the original consumed resources is determined as the historical adjustment coefficient of the target information within the first preset time period.

In this embodiment, the value of each historical resource is weighted and averaged, and the ratio of the obtained average value of historical resources to the original consumed resources is used to determine the historical adjustment coefficient, so that the historical adjustment coefficient can be accurately determined according to the correlation between the average historical resource and the original consumed resources Determine the historical adjustment factor over the past time period.

In one embodiment, the first determining module 904 is further configured to: obtain the historical adjustment coefficient of the target information within the first preset time period, and the corresponding consumed resources; The unadjusted expected consumption resource within the first preset time period.

In this embodiment, by adjusting the value of the candidate resource of the target information, the historical adjustment coefficient of the target information in the first preset duration, and the average cost of obtaining one conversion, it is possible to accurately calculate the value of the candidate resource for the target information. In the case where the resource value is not adjusted, the target information obtains the average cost of one conversion in the first preset duration, so that the conversion cost in the case of adjustment can be accurately predicted by the conversion cost in the case of adjustment.

In one embodiment, the first determining module 904 is further configured to: obtain a historical adjustment coefficient of the target information within a first preset time period, and determine, based on the accumulated consumed resources and the historical adjustment coefficient, to determine the first preset time period when the target information is not adjusted. Set the unadjusted expected cumulative consumption resource corresponding to the duration; obtain the smoothing factor corresponding to the unadjusted expected consumption resource; according to the original consumption resource, the unadjusted expected consumption resource, the smoothing factor, the predicted adjustment coefficient and the The expected accumulated consumption resources are determined, and the adjustment expected consumption resources corresponding to the second preset time period when the target information is adjusted is determined.

In this embodiment, according to the historical adjustment coefficient and the accumulated consumption resources in the past time period, the expected accumulated consumption of the unadjusted price in the past time period is accurately calculated under the assumption that the target information is not adjusted. The unadjusted expected accumulated consumption error in the past time period is relatively large, and the error can be reduced by the smoothing factor. Based on various factors such as original consumption resources, unadjusted expected consumption resources, smoothing factors, predicted adjustment coefficients, and unadjusted expected cumulative consumption resources, etc., estimate the adjustment expected consumption resources of target information in the future time period, and consider the data More comprehensive and more accurate estimates.

In one embodiment, the first determining module 904 is further configured to: determine the predicted adjustment coefficient, the first product of the unadjusted expected cumulative consumption resource corresponding to the first preset duration and the unadjusted expected consumption resource, and the predicted adjustment coefficient, smoothing factor, the second product of the unadjusted expected consumption resource and the original consumption resource; determine the third product of the smoothing factor and the unadjusted expected consumption resource and the sum of the unadjusted expected accumulated consumption resource corresponding to the first preset duration; Multiplying by the second product and dividing by the third product and the sum of the unadjusted expected cumulative consumption resources corresponding to the first preset duration, obtains the adjusted expected consumption resources corresponding to the second preset duration when the target information is adjusted.

In this embodiment, according to the known target information in the past time period, unadjusted expected consumption resources, unadjusted expected accumulated consumption resources, smoothing factor, and the original consumption resources set by the information owner, it can be accurately predicted that in the future time period The adjustment of target information is expected to consume resources. According to the known information in the past time period, the resource consumption of the target information in the future time period is estimated, so as to fully plan the push of the target information in the future time period, so as to realize the accurate push of the target information.

In one embodiment, the smoothing factor is obtained through a smoothing factor determining step, and the smoothing factor determining step includes: acquiring the historical conversion amount corresponding to the target information in the historical duration adjacent to the target duration; Set the duration and historical conversion amount to obtain the smoothing factor corresponding to the unadjusted expected consumption resource.

In this embodiment, by calculating a smoothing factor, the adjustment expected resource consumption of the target information within the first preset time period is smoothed by the smoothing factor, so as to smooth the estimated consumption resource of the target information in the past time period, It can reduce the error of estimated resource consumption.

In one embodiment, the second determining module 906 is further configured to: determine the sum of the accumulated consumption resources and the adjusted expected accumulated consumption resources, as well as the sum of the converted amount and the expected conversion amount; compare the accumulated consumption resources with the adjusted expected cumulative consumption The sum of resources, and the ratio of the converted amount to the expected conversion amount, is determined as the target resource consumption of the target information within the target duration.

In this embodiment, by using the relationship between the converted amount, the accumulated consumption resources, the expected conversion amount, and the adjusted expected cumulative consumption resources, it is predicted that the target information can be obtained within the target duration when the candidate resource value selected this time is used. Transform the resource expected to be consumed at one time, so as to determine whether the resource expected to be consumed is eligible to perform the corresponding operation. If the condition is met, it means that accurate push of information can be achieved based on the value of the candidate resource selected in this round. If the condition is not met, it means that the value of the candidate resource selected in this round cannot achieve accurate push of information, and the next round of selection of the candidate resource value will be performed. and calculate.

In one embodiment, the pushing module 908 is further configured to: when the target consumption resource and the original consumption resource do not meet the information pushing conditions, select another candidate resource value from the candidate set, and return to execute the determination that the selected candidate resource value corresponds to The step of predicting the adjustment coefficient is to obtain the target consumption resource of the target information within the target duration; when the target consumption resource and the original consumption resource meet the information push conditions, the target information is pushed according to the selected candidate resource value.

In this embodiment, the candidate resource values are selected by traversal from the candidate set, and the target consumption resources consumed by one conversion within the target duration are calculated based on the selected candidate resource values. The value of the candidate resource, so as to realize the accurate push of information according to the value of the candidate resource.

In one embodiment, the pushing module 908 is further configured to: determine the corresponding expected income of each target information when the information is pushed according to the respective candidate resource values; and push the target information corresponding to the expected income satisfying the income condition.

In this embodiment, the expected income of the target information is used as a condition for information push, and the target information corresponding to the expected income satisfying the income condition is pushed, so as to achieve accurate information push.

In one embodiment, the target information is a target advertisement, the original consumption resource is the original conversion unit price, the candidate resource value is the candidate bid price, the accumulated consumption resource is the converted total price, and the adjusted expected accumulated consumption resource is the price adjustment expected conversion total price, The target consumption resource is the target conversion unit price;

The push module 908 is further configured to: when the target conversion unit price is the same as the original conversion unit price, determine the advertisement placement position corresponding to the candidate bid; and place the target advertisement at the advertisement placement position within the second preset time period.

In this embodiment, the converted amount and cost of the target advertisement by users in the past time period are fully considered, and according to the converted amount, cost, historical price adjustment coefficient and other factors in the past time period, it is accurately estimated that the user's interest in the target advertisement in the future time period is accurately estimated. The estimated conversion volume and cost of the targeted ad. And based on the user's conversion volume and cost of the advertisement in the past time period, as well as the user's estimated conversion volume and cost of the target advertisement in the future time period, more accurately predict the target advertisement's conversion cost in the future time period. , so as to more accurately determine the price of target advertising. At the same time, if the target conversion unit price and the original conversion unit price satisfy the advertisement placement conditions, it means that the predicted cost of conversion of the target advertisement in the future time period is roughly the same as the cost defined by the advertiser, and the corresponding advertisement is placed according to the candidate bid. Positioning targeted advertisements can achieve accurate advertisement placement, thereby obtaining higher advertisement conversions and improving the efficiency of advertisement placement under the condition of meeting the cost provided by the advertisement owner.

In one embodiment, the first preset duration is the duration from the starting point of the target duration to the current moment, the second preset duration is the duration from the current moment to the end of the target duration, and the target duration is a periodic fixed duration.

In this embodiment, the target duration is a periodic fixed duration, and the information push method can be based on the actual situation of the target information within each target duration, such as the converted amount, accumulated consumption resources and other candidate resources for the target information The value is adjusted in time, so that the target information is pushed with the appropriate candidate resource value, and the accurate push of the target information is automatically realized.

For the specific limitation of the information pushing apparatus, please refer to the limitation on the information pushing method above, which will not be repeated here. Each module in the above-mentioned information pushing device can be implemented in whole or in part by software, hardware and combinations thereof. The above modules can be embedded in or independent of the processor in the computer device in the form of hardware, or stored in the memory in the computer device in the form of software, so that the processor can call and execute the operations corresponding to the above modules.

In one embodiment, a computer device is provided, and the computer device may be a terminal, and its internal structure diagram may be as shown in FIG. 10 . The computer equipment includes a processor, memory, a communication interface, a display screen, and an input device connected by a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium, an internal memory. The nonvolatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the execution of the operating system and computer programs in the non-volatile storage medium. The communication interface of the computer device is used for wired or wireless communication with an external terminal, and the wireless communication can be realized by WIFI, operator network, NFC (Near Field Communication) or other technologies. When the computer program is executed by the processor, an information push method is implemented. The display screen of the computer equipment may be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer equipment may be a touch layer covered on the display screen, or a button, a trackball or a touchpad set on the shell of the computer equipment , or an external keyboard, trackpad, or mouse.

Those skilled in the art can understand that the structure shown in FIG. 10 is only a block diagram of a partial structure related to the solution of the present application, and does not constitute a limitation on the computer equipment to which the solution of the present application is applied. Include more or fewer components than shown in the figures, or combine certain components, or have a different arrangement of components.

In one embodiment, a computer device is also provided, including a memory and a processor, where a computer program is stored in the memory, and the processor implements the steps in the foregoing method embodiments when the processor executes the computer program.

In one embodiment, a computer-readable storage medium is provided, which stores a computer program, and when the computer program is executed by a processor, implements the steps in the foregoing method embodiments.

In one embodiment, there is provided a computer program product or computer program comprising computer instructions stored in a computer readable storage medium. The processor of the computer device reads the computer instructions from the computer-readable storage medium, and the processor executes the computer instructions, so that the computer device executes the steps in the foregoing method embodiments.

Those of ordinary skill in the art can understand that all or part of the processes in the methods of the above embodiments can be implemented by instructing relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage In the medium, when the computer program is executed, it may include the processes of the above-mentioned method embodiments. Wherein, any reference to memory, storage, database or other media used in the various embodiments provided in this application may include at least one of non-volatile and volatile memory. The non-volatile memory may include Read-Only Memory (ROM), magnetic tape, floppy disk, flash memory or optical memory, and the like. Volatile memory may include random access memory (RAM) or external cache memory. By way of illustration and not limitation, the RAM may be in various forms, such as static random access memory (Static Random Access Memory, SRAM) or dynamic random access memory (Dynamic Random Access Memory, DRAM).

The technical features of the above embodiments can be combined arbitrarily. In order to make the description simple, all possible combinations of the technical features in the above embodiments are not described. However, as long as there is no contradiction in the combination of these technical features It is considered to be the range described in this specification.

The above-mentioned embodiments only represent several embodiments of the present application, and the descriptions thereof are specific and detailed, but should not be construed as a limitation on the scope of the invention patent. It should be pointed out that for those skilled in the art, without departing from the concept of the present application, several modifications and improvements can be made, which all belong to the protection scope of the present application. Therefore, the scope of protection of the patent of the present application shall be subject to the appended claims.

Claims (15)

1. an information push method, is characterized in that, described method comprises: Obtain the original consumption resource and the candidate resource value corresponding to the target information, and determine the prediction adjustment coefficient corresponding to the candidate resource value; Obtain the converted amount and accumulated consumption resources of the target information within a first preset time period, and determine the adjustment of the target information within a second preset time period based on the predicted adjustment coefficient and the accumulated consumption resources Expected cumulative consumption of resources and expected conversion; According to the converted amount, the accumulated consumption resources, the expected conversion amount, and the adjusted expected accumulated consumption resources, determine the target consumption resources of the target information within the target duration; the target duration includes the first 2. a preset duration and the first preset duration; When the target consumption resource and the original consumption resource meet the information push condition, the target information is pushed according to the candidate resource value. 2. The method according to claim 1, characterized in that, obtaining the corresponding original consumption resources and candidate resource values of the target information, and determining the prediction adjustment coefficient corresponding to the candidate resource values, comprising: Obtain the original consumption resource corresponding to the target information, and select the candidate resource value for pushing the target information from the candidate set; A prediction adjustment coefficient corresponding to the candidate resource value is determined according to the candidate resource value and the original consumed resource. 3 . The method according to claim 1 , wherein, based on the predicted adjustment coefficient and the accumulated consumption resources, determining the adjustment expected accumulated consumption resources of the target information within the second preset time period and the Expected conversions, including: determining, based on the accumulated consumption resources, the corresponding adjustment expected accumulated consumption resources within the second preset time period when the target information is adjusted; When acquiring the target information that is not adjusted, the corresponding unadjusted expected consumption resource within the first preset time period; Determine, based on the predicted adjustment coefficient and the unadjusted expected consumption resource, the adjustment expected consumption resource corresponding to the second preset time period when the target information is adjusted; According to the adjusted expected accumulated consumption resource and the adjusted expected consumption resource, the expected conversion amount of the target information within the second preset time period is determined. 4 . The method according to claim 3 , wherein, when the target information is adjusted based on the accumulated consumption resources, the corresponding adjustment expected accumulated consumption resources within a second preset time period are determined, comprising: 4 . Acquire the historical adjustment coefficient of the target information within the first preset time period, and based on the accumulated consumed resources and the historical adjustment coefficient, determine that the target information is within the first preset time period when the target information is not adjusted The corresponding unadjusted expected cumulative consumption resources; Based on the first preset duration and the unadjusted expected accumulated consumption resource corresponding to the first preset duration, determining the unadjusted expected accumulated consumption resource corresponding to the second preset duration when the target information is not adjusted; Based on the predicted adjustment coefficient and the unadjusted expected cumulative consumption resource corresponding to the first preset time period, determine the adjustment expected cumulative consumption resource corresponding to the second preset time period when the target information is adjusted. 5 . The method according to claim 4 , wherein, based on the first preset duration and the unadjusted expected cumulative consumption resources corresponding to the first preset duration, determining when the target information is unadjusted The unadjusted expected cumulative consumption resources corresponding to the second preset duration include: According to the first preset duration and the unadjusted expected accumulated consumption resource corresponding to the first preset duration, the expected accumulated consumption resource of the target information within the target duration is determined; the target duration includes the first expected accumulated consumption resource. set duration and second preset duration; Based on the expected cumulative consumption resources within the target duration and the unadjusted expected cumulative consumption resources corresponding to the first preset duration, determine the unadjusted corresponding to the second preset duration when the target information is not adjusted Expect a cumulative consumption of resources. 6. The method according to claim 4, wherein the acquiring the historical adjustment coefficient of the target information within the first preset time period comprises: Obtaining the historical resource value used for pushing the target information within the first preset duration; According to the historical resource value and the original consumed resource, a historical adjustment coefficient of the target information within the first preset time period is determined. 7 . The method according to claim 3 , wherein when the acquisition of the target information is not adjusted, the corresponding unadjusted expected consumption resources within the first preset time period comprise: 7 . acquiring the historical adjustment coefficient of the target information within the first preset time period, and the corresponding consumed resources; According to the consumed resource and the historical adjustment coefficient, the unadjusted expected consumption resource of the target information within the first preset time period is determined. 8 . The method according to claim 3 , wherein, based on the predicted adjustment coefficient and the unadjusted expected consumption resource, determining the target information corresponding to the second preset time period when the target information is adjusted. 9 . The conditioning expectation consumes resources, including: Acquire the historical adjustment coefficient of the target information within the first preset time period, and based on the accumulated consumed resources and the historical adjustment coefficient, determine that the target information is within the first preset time period when the target information is not adjusted The corresponding unadjusted expected cumulative consumption resources; obtaining the smoothing factor corresponding to the unadjusted expected consumption resource; According to the original consumption resource, the unadjusted expected consumption resource, the smoothing factor, the predicted adjustment coefficient, and the unadjusted expected accumulated consumption resource corresponding to the first preset duration, it is determined that the target information is adjusted when The corresponding adjustment within the second preset time period is expected to consume resources. 9. The method according to claim 8, wherein the smoothing factor is obtained through a smoothing factor determining step, and the smoothing factor determining step comprises: Obtain the historical conversion amount corresponding to the target information in the historical duration adjacent to the target duration; Based on the target duration, the first preset duration and the historical conversion amount, a smoothing factor corresponding to the unadjusted expected consumption resource is obtained. 10. The method according to any one of claims 1 to 9, wherein the method further comprises: When the target consumption resource and the original consumption resource do not meet the information push conditions, select another candidate resource value from the candidate set, and return to the step of determining the prediction adjustment coefficient corresponding to the selected candidate resource value, so as to obtain the Describe the target consumption resources of the target information within the target duration; When the target consumption resource and the original consumption resource meet the information push condition, the target information is pushed according to the selected candidate resource value. 11. The method according to any one of claims 1 to 9, wherein the method further comprises: Determine the corresponding expected income of each target information when the information is pushed according to the respective candidate resource values; Push the target information corresponding to the expected income that meets the income conditions. 12. The method according to any one of claims 1 to 9, wherein the target information is a target advertisement, the original consumption resource is an original conversion unit price, the candidate resource value is a candidate bid, and the The accumulated consumption resource is the converted total price, the adjusted expected accumulated consumption resource is the price adjusted expected conversion total price, and the target consumption resource is the target conversion unit price; When the target consumption resource and the original consumption resource meet the information push conditions, the target information is pushed according to the candidate resource value, including: When the target conversion unit price is the same as the original conversion unit price, determine the advertisement placement position corresponding to the candidate bid; The target advertisement is placed at the advertisement placement position within the second preset time period. 13. An information push device, characterized in that the device comprises: an acquisition module, configured to acquire the original consumption resource and the candidate resource value corresponding to the target information, and determine the prediction adjustment coefficient corresponding to the candidate resource value; The first determination module is configured to obtain the converted amount and the accumulated consumption resources of the target information within a first preset time period, and determine that the target information is in the first 2. Adjustment of expected cumulative consumption of resources and expected conversion within a preset time period; a second determining module, configured to determine the target consumption resource of the target information within the target duration according to the converted amount, the accumulated consumption resource, the expected conversion amount and the adjusted expected cumulative consumption resource; The target duration includes the second preset duration and the first preset duration; A push module, configured to push the target information according to the candidate resource value when the target consumption resource and the original consumption resource meet the information push condition. 14. A computer device, comprising a memory and a processor, wherein the memory stores a computer program, wherein the processor implements the method according to any one of claims 1 to 12 when the processor executes the computer program. step. 15. A computer-readable storage medium storing a computer program, wherein when the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 12 are implemented.

Images