KR20200029821A - Providing Method of parameter and server device supporting the same - Google Patents

Abstract

The present invention discloses a method for extracting a parameter based on and a server device supporting the same. The server device may include a communication circuit, a memory, and a processor functionally connected to the communication circuit. The processor calculates similarity scores between categories of a plurality of advertisements based on user data, and calculates user scores for categories for each user terminal using at least one advertisement among the plurality of advertisements based on the similarity scores. , It can be set to calculate the parameters for each category based on the user scores, and to store the calculated parameters for each category.

Description

{Providing Method of parameter and server device supporting the same}

The present invention relates to a parameter extraction method capable of extracting parameters for sending advertisements in a more logical manner in relation to parameter extraction, and a server device supporting the parameter extraction.

With the development of mobile communication networks and the development of terminal specifications, mobile communication terminals have become essential belongings of modern people beyond the scope of conventional simple communication devices or information providing devices, and are evolving into total entertainment devices. In particular, as the use of the Internet has become more common in recent years, users have been able to acquire various information through Internet search. That is, after accessing the Internet search site through a terminal device such as a personal computer that can access the Internet, users can search for various contents related to news, knowledge, games, and communities.

As such, as information acquisition through Internet search has become common, recently, a method for providing advertisements necessary for a user has been sought in response to an action log or user input for advertisements provided to the user. In particular, appropriate parameter extraction is necessary to provide effective advertisements to various users.

Considering the case of conventional parameter extraction, for example, in the case of automobile advertisements, there are many cases in which a user has seen an automobile driver or extracts parameters using an intuitive method of an advertisement manager or an advertisement planner. However, if there are many advertisements or if the advertisements continue to progress, there is a limit to relying on an advertisement planner.

The present invention applies various logics used in recommendation to obtain advertisement category similarity, and additionally obtains a similarity score between a user and a category and between a category and a user to use a customer with a high score as a parameter when executing an advertisement, It is intended to provide a parameter extraction method for an advertisement service capable of providing advertisement data according to a more reliable parameter and a server device supporting the parameter extraction method.

The server device according to an embodiment of the present invention is to achieve the above object, the server device according to an embodiment of the present invention is a communication circuit, at least one collected from at least one user terminal connected through the communication circuit It may include a memory for storing user data related to the use of advertisements by category, and the communication circuit and a processor functionally connected to the memory. The processor calculates similarity scores between categories of a plurality of advertisements based on the user data, and calculates user scores for categories for each user terminal using at least one advertisement among the plurality of advertisements based on the similarity scores And, it can be set to calculate the parameters for each category based on the user scores, and to store the calculated parameters for each category.

In addition, the processor may be configured to determine the ranking of categories by user terminal using the user scores, and to calculate the parameters for each category using the ranking of categories by user terminal.

Also, the processor may filter and remove data that will have a similarity of 0 among data included in the user data.

In particular, the processor may be configured to calculate similarity scores among a plurality of advertisement categories of the user data using a cosine similarity algorithm or a jacquard similarity algorithm.

Meanwhile, in relation to a specific user terminal, the processor multiplies the use history values of advertisements of a plurality of categories by multiplying similarity scores between the categories as a molecular value, and the sum of the similarity scores between the categories is a denominator value. A value may be set as the user score of the specific user terminal.

The processor may set index values of user terminals, which are prioritized for each of the plurality of advertisement categories, as parameters for each category.

Alternatively, the processor may be configured to determine a specific user terminal that transmits an advertisement of a specific category by checking the parameters for each category.

Here, the processor may be configured to transmit the advertisement of the specific category based on the parameter of the next subordinate priority when there is no priority parameter of the specific category.

The method for extracting parameters of a server device according to an embodiment of the present invention includes collecting user data related to advertisement use for at least one category from at least one user terminal connected to the server device through a communication circuit, and collecting the user data. Calculating similarity scores between categories of a plurality of advertisements based on the results; calculating user scores for categories for each user terminal using at least one advertisement among the plurality of advertisements based on the similarity scores; and the user score And calculating a parameter for each category based on them.

In addition, the method may include determining the ranking of categories for each user terminal using the user scores, filtering and removing data having similarity 0 among data included in the user data, or removing parameters for each category. It may further include at least one step of checking and transmitting the advertisement of a specific category to a specific user terminal.

Here, in the calculating of the user score, in relation to a specific user terminal, a multiplication value is multiplied by multiplying similarity scores between the categories to a usage history value for each advertisement as a molecular value, and the similarity score between the categories It may include the step of calculating the value of the sum of the denominator value as the user score of the specific user terminal.

According to the present invention, by using recommendation logic in a method of performing parameter extraction for advertisement execution, it is possible to extract parameters through a clear definition based on data. Can be further increased.

Other, other effects of the present invention may be provided by various embodiments described below.

1 is a diagram illustrating an example of an environment to which an advertisement management system according to an embodiment of the present invention is applied.
2 is a view showing an example of a server configuration of a server device according to an embodiment of the present invention.
3 is a view showing an example of a signal flow between each component of the advertisement management system according to an embodiment of the present invention.
4 is a view showing an example of a server device configuration according to an embodiment of the present invention.
5 is a diagram illustrating an example of a processor configuration according to an embodiment of the present invention.
6 is a diagram illustrating an example of a method of operating a server device related to an advertisement operation according to an embodiment of the present invention.

Hereinafter, preferred embodiments of the present invention will be described in detail with reference to the accompanying drawings. However, in the following description and accompanying drawings, detailed descriptions of well-known functions or configurations that may obscure the subject matter of the present invention are omitted. In addition, it should be noted that the same components throughout the drawings are indicated by the same reference numerals as much as possible.

The terms or words used in the present specification and claims described below should not be construed as being limited to ordinary or dictionary meanings, and the inventor is appropriate as a concept of terms for explaining his or her invention in the best way. It should be interpreted as a meaning and a concept consistent with the technical idea of the present invention based on the principle that it can be defined as such. Therefore, the configuration shown in the embodiments and drawings described in this specification is only one of the most preferred embodiments of the present invention, and does not represent all of the technical spirit of the present invention, and can replace them at the time of this application. It should be understood that there may be equivalents and variations.

Hereinafter, a terminal according to an embodiment of the present invention will be described as a representative example of a mobile communication terminal capable of executing an application connected to a communication network and receiving advertisements through the application, but the terminal is not limited to a mobile communication terminal, and all information It can be applied to various terminals such as communication devices, multimedia terminals, wired terminals, fixed terminals, and IP (Internet Protocol) terminals. In addition, the terminal is a mobile phone, a portable multimedia player (PMP), a mobile Internet device (MID), a smartphone (Smart Phone), a desktop (Desktop), a tablet computer (Tablet PC), a notebook (Note book), a netbook (Net book) And mobile terminals having various mobile communication specifications, such as information and communication equipment.

1 is a view showing an example of an environment to which an advertisement management system is applied according to an embodiment of the present invention, and FIG. 2 is a view showing an example of a server configuration of a server device according to an embodiment of the present invention.

1 and 2, the advertisement management system 10 according to an embodiment of the present invention provides an advertisement to the user terminal 100 and the user terminal 100 including at least one terminal and feedback accordingly After collecting, the advertisement parameter may be extracted and operated to include the server device 200 and the network 50. Here, the server device 200 may include a data collection server 201, a recommendation engine server 203, and an advertisement server 205.

The network 50 to which the present invention is applied is not only an IP-based wired communication network such as an Internet network, but also various types of wireless networks such as a long term evolution (LTE) network, a mobile communication network such as a WCDMA network, and a Wi-Fi network. , And combinations thereof. That is, the advertisement operation system according to the present invention can be applied to all wired and wireless communication networks without distinction. Specifically, the network 50 may form a communication channel between the server device 200 and the user terminal 100. For example, the network 50 may support at least one of 3G, 4G, and 5G wireless mobile communication methods that can be operated by the server device 200 or the user terminal 100. Alternatively, the network 50 may be a wire-based form a communication channel of a communication channel between the user equipment 100 and the server 200. Such a network 50 should be interpreted as a concept including various wired networks, wireless networks, and combination networks thereof which are currently developed and commercialized or will be developed and commercialized in the future. The data collection server 201, the recommendation engine server 203 and the advertisement server 205 constituting the server device 200 may be connected through the network 50. When the server device 200 includes a data collection server 201, a recommendation engine server 203, and an advertisement server 205, the user terminal 100 includes a data collection server 201 and an advertisement server 205. On the network 50 can be connected. Alternatively, the data collection server 201, the recommendation engine server 203, and the advertisement server 205 constituting the server device 200 may be connected through separate cables.

The user terminal 100 forms a communication channel with the server device 200 through the network 50, or at least one of the data collection server 201 and the advertisement server 205 included in the server device 200. Can be connected. For example, the user terminal 100 may establish a communication channel with the advertisement server 205 and receive advertisements belonging to at least one category from the advertisement server 205. The user terminal 100 outputs at least one advertisement provided by the advertisement server 205, and transmits a response signal corresponding to advertisement selection to the data collection server 201 in response to an action log (eg, user input) Can. In this regard, the user terminal 100 may store an application that can access the advertisement server 205 or an application, such as a web browser capable of driving a website where the advertisement server 205 provides advertisements. The above-described user terminal 100 may include various terminals each of a plurality of users. For example, the user terminal 100 is an electronic device capable of outputting advertisements, and may be various devices such as a smartphone or a tablet PC, a desktop PC, and a camera equipped with a communication circuit.

The above-described user terminal 100 may include, for example, a communication circuit, an input device, a memory, a display, and a processor. The communication circuit of the user terminal 100 forms a communication channel with the network 50, and based on the network 50, data related to viewing advertisements (eg, advertisements belonging to at least one category) of the server device 200 It can be received from the advertisement server 205. For example, the communication circuit forms a communication channel with a server (eg, the advertisement server 205) having address information designated according to the execution of the web browser, and receives at least one advertisement belonging to a specific category from the advertisement server 205. You can. Further, the communication circuit may transmit an action log (eg, user input) in connection with the received advertisement to the data collection server 201 of the server device 200. The communication circuit may form a communication channel based on a long-range wireless communication network such as 3G, 4G, or 5G, or a communication channel based on a short-range communication network such as Wi-Fi.

The input device of the user terminal 100 may receive user input related to manipulation of the user terminal 100. For example, the input device may be a user input for executing a web browser, a text input (for example, address information) required for accessing the advertisement server 205, or a button key or virtual button key for selecting an application execution icon, voice It can support input functions, etc. After receiving the advertisement from the advertisement server 205, the input device may transmit an input signal according to a user manipulation to the processor of the user terminal 100. The input signal input through the input device may be transmitted to the data collection server 201 in response to processor control. Alternatively, operation information related to viewing a selected or viewed advertisement corresponding to the input signal may be transmitted to the data collection server 201 in response to processor control.

The memory of the user terminal 100 may store data or application programs related to the operation of the user terminal 100. For example, the memory may store a program (or application) related to advertisement viewing or a web browser supporting an advertisement viewing function. The memory may store at least one advertisement received from the advertisement server 205, and corresponding to the input signal from the input device, the stored at least one advertisement may be output on the display of the user terminal 100.

The display of the user terminal 100 may output at least one screen related to the operation of the user terminal 100. For example, the display may output an execution screen according to the execution of the advertisement viewing function, a connection screen of the server device 200 (eg, an advertisement server 205 connection screen), and the like.

The processor of the user terminal 100 may perform processing or transmission of data related to the operation of the user terminal 100. For example, the processor of the user terminal 100 may execute an advertisement viewing program (eg, a specific web browser or web application) in response to a user input input through an input device. In this regard, the processor of the user terminal 100 forms a communication channel with the server device 200 (eg, the advertisement server 205) using the communication circuit of the user terminal 100, and the advertisement server ( 205), and output the advertisement. When the processor of the user terminal 100 outputs an advertisement of a specific category on a display and receives an action log (eg, user input) requesting selection or reproduction of an advertisement of a specific category, the advertisement requested to be selected or played It is possible to transmit the received user input to the data collection server 201 while playing.

The data collection server 201 may collect and store an action log from the user terminal 100. In this regard, the data collection server 201 may form a communication channel with the user terminal 100. For example, the data collection server 201 obtains connection information of the user terminal 100 providing advertisement of a specific category from the advertisement server 205, and based on the obtained connection information, the user terminal 100 and A communication channel can be formed. The data collection server 201 may collect advertisement-related user input information or advertisement-related control information generated in the user terminal 100. The data collection server 201 may receive the user input information or the advertisement-related control information as user data 210 and store it. According to various examples, the data collection server 201 may establish a communication channel with the advertisement server 205 and collect user data 210 of the user terminal 100 through the advertisement server 205. For example, when the advertisement server 205 provides an advertisement of a specific category to the user terminal 100, and transmits control information related to a specific category reproduction control to the user terminal 100, the advertisement server 205, the data collection server 201 may collect advertisement reproduction control information related to the user terminal 100 received by the advertisement server 205 as user data 210. The data collection server 201 may deliver the collected user data 210 to the recommendation engine server 203 when the user data 210 is collected in a specified amount or for a specified time. The user data 210 collected by the data collection server 201 may include information on advertisement categories provided to a plurality of user terminals and information on advertisement categories viewed by a plurality of user terminals.

The recommendation engine server 203 may provide parameters for each advertisement so that similarities between items may be obtained through analysis of collected data and used in the advertisement server 205. In this regard, the recommendation engine server 203 may establish a communication channel with the data collection server 201 and collect user data 210 from the data collection server 201. In this operation, the recommendation engine server 203 may establish a communication channel with the data collection server 201 according to a specified time period, and collect user data 210 from the data collection server 201. The recommendation engine server 203 may calculate a similarity value for the user data 210 using a recommendation algorithm specified for the received user data 210. For example, the recommendation engine server 203 may calculate a similarity value between a user and advertisement categories or a similarity value between categories for the user data 210 using a jacquard similarity algorithm. When the similarity value is calculated, the recommendation engine server 203 may transmit a score of a user having a relatively similarity value to the advertisement server 205 so that it can be used as a parameter in advertisement execution. In this regard, the recommendation engine server 203 may include at least one recommendation logic 230. The recommendation engine server 203 acquires data on the viewing history of each user's advertisement category or the purchase history for each advertisement category from the user data 210, and at least one of various recommendation logics 230 in the obtained user data 210 A similarity table between categories may be generated by applying one logic. The recommendation engine server 203 may acquire a user score between users and categories, and use a relatively largest score among user scores as a parameter of the advertisement. The recommendation logic 230 is a Jacquard similarity calculation algorithm of dividing the size of the intersection of two data sets by the size of the union, and the degree of similarity between the vectors measured using the cosine value of the angle between the two vectors in the inner space. It may include a cosine similarity algorithm of calculating a method, a Pearson similarity algorithm of measuring a trend through a one-to-one comparison of two consecutive data. In addition to the jacquard similarity algorithm, cosine similarity algorithm, and Pearson similarity algorithm described above, various logics for measuring similarity may be applied to the recommended logic 230 of the present invention.

The advertisement server 205 may transmit advertisements based on parameters for each advertisement provided by the recommendation engine server 203. In this regard, the advertisement server 205 may include an advertisement DB 250. When the advertisement delivery time arrives, or when a specific user function execution or specific application execution of a specific user terminal 100 is performed, when the advertisement is requested, the advertisement server 205 recommends the engine The server 203 may request parameter values for each advertisement category. When the advertisement server 205 receives a parameter value for each advertisement category from the recommendation engine server 203, the advertisement server 205 may transmit an advertisement of a specific category to a specific user terminal based on the received parameter value. The advertisement DB 250 may include advertisements of various categories. Alternatively, the advertisement DB 250 may include product information related to various products. The advertisement server 205 may select advertisements of a specific category based on the parameters provided by the recommendation engine server 203 and transmit advertisements of the selected category to at least one user terminal 100.

3 is a view showing an example of a signal flow between each component of the advertisement management system according to an embodiment of the present invention.

Referring to FIG. 3, in the advertisement management system of the present invention, in step 301, at least one user terminal 100 may transmit user data as a data collection server 201 as an action log. In this regard, the advertisement server 205 may form a communication channel with the user terminal 100 and deliver a specific category advertisement according to the policy to the user terminal 100. In this operation, the advertisement server 205 may deliver at least one advertisement belonging to a specific category to the user terminal 100 or may transmit advertisements for each category to the user terminal 100. The user terminal 100 may perform an operation such as requesting to play at least one advertisement provided by the advertisement server 205 or purchasing a product through the advertisement. In this regard, the user terminal 100 may generate a user input for selecting a specific advertisement through an input device, and transmit the user input to the data collection server 201 as an action log of the user terminal 100. The data collection server 201 may store user data (eg, advertisement information selected by the user input) corresponding to the action log of the user terminal 100 in the memory.

In step 303, the data collection server 201 may perform user data collection classification. For example, the data collection server 201 may classify and store collected user data by user (or user terminal) and category.

In step 305, the data collection server 201 may transmit the user data 210 to the recommendation engine server 203. For example, the data collection server 201 may deliver the collected user data 210 to the recommendation engine server 203 when the designated time arrives. Alternatively, the data collection server 201 may deliver the corresponding user data 210 to the recommendation engine server 203 when the amount of the collected user data 210 becomes a specified number. Alternatively, the data collection server 201 recommends the user data 210 collected according to the request of the advertisement server 205 when the advertisement server 205 is ready to transmit advertisements of a specified category. 203).

In step 307, the recommendation engine server 203 may perform parameter extraction. For example, the recommendation engine server 203 may calculate the similarity value of a designated comparison criterion for the user data 210 using at least one recommendation logic among the stored recommendation logics 230. For example, the recommendation engine server 203 may calculate a similarity score between categories of advertisements, or a similarity score between users or a similarity score between users and categories. The recommendation engine server 203 may set a user score based on the calculated similarity score. The recommendation engine server 203 may extract a parameter based on the established user score.

In step 309, the recommendation engine server 203 may deliver the extracted parameters to the advertisement server 205. The recommendation engine server 203 may select, as a parameter, the score of the user with the highest user score relative to the advertisement of a specific category, and transmit it to the advertisement server 205.

In step 311, the advertisement server 205 may perform advertisement execution based on the received parameters. For example, the advertisement server 205 may transmit the advertisement to a user terminal having a high parameter.

Meanwhile, in the above description, the server device 200 has been described based on a form including a data collection server 201, a recommendation engine server 203, and an advertisement server 205, but the present invention is not limited thereto. For example, the server device 200 may be configured as one independent electronic device in addition to the form including the configuration of each of the servers described above. In this case, the server device 200 may include a configuration as shown in FIG. 4.

4 is a view showing an example of a server device configuration according to an embodiment of the present invention.

Referring to FIG. 4, the server device 200 (or server device) according to an embodiment of the present invention may include a communication circuit 210, a memory 230, and a processor 250. The memory 230 may be provided in a configuration independent of the server device 200. In this case, the processor 250 may control to form a communication channel with a separately configured memory 230 device or server, and to perform a search operation.

The communication circuit 210 may be in charge of the communication function of the server device 200. That is, the communication circuit 210 may form a communication channel with at least one user terminal 100 through the network 50. The communication circuit 210 may transmit at least one advertisement, receive user data 210 from the user terminal 100, and the like, in response to control of the processor 250.

The memory 230 may store data or programs related to the operation of the server device 200. For example, the memory 230 may store user data 210, recommendation logic 230, and advertisement DB 250 in connection with providing an advertisement service.

The user data 210 may include action logs for advertisements received from a plurality of user terminals 100. For example, the user data 210 may include information indicating whether a user is a user with respect to a specific category advertisement provided to a specific user terminal 100, or information indicating whether a product has been purchased based on a specific category advertisement. have. The user data 210 may be used to calculate parameters for sending advertisements.

The recommendation logic 230 may include at least one logic used to calculate a parameter used to transmit advertisements based on the user data 210. For example, the recommendation logic 230 may include a cosine similarity algorithm, a jacquard similarity algorithm, and the like. The recommendation logic 230 may be used to calculate a similarity score between categories of advertisements based on the user data 210, or may be used to calculate a score regarding relationships between users and categories.

The advertisement DB 250 may store at least one advertisement to be provided to the user terminal 100. The advertisements stored in the advertisement DB 250 may be classified into certain categories, for example. As an example, advertisements stored in the advertisement DB 250 may be divided into a large category category, a middle category category, and a small category category and stored. The advertisement stored in the advertisement DB 250 may include, for example, advertisements related to various product guides or product purchases. For example, the advertisement may guide various products that can be sold online, such as shoes, shoes, clothes, electronic products, food, and travel products, and may include information that can be purchased by user selection.

The processor 250 may perform data processing or program operation related to the operation of the server device 200. For example, the processor 250 supports the connection of the user terminal 100, provides at least one advertisement to the user terminal 100, and watch an advertisement or a corresponding advertisement from the user terminal 100 that provided the advertisement. The action log related to product purchase based on the user data 210 may be collected. The processor 250 may calculate a user score corresponding to a specific category advertisement based on the collected user data 210 and extract parameters based on the user score. The processor 250 may provide advertisements of a specific category to the designated user terminal 100 based on the calculated parameters. In this regard, the processor 250 may include a configuration as shown in FIG. 5.

5 is a diagram illustrating an example of a processor configuration according to an embodiment of the present invention.

Referring to FIG. 5, the processor 250 of the present invention may include an information collection unit 251, a parameter extraction unit 253, and an advertisement execution unit 255.

The information collecting unit 251 may collect user data 210. In this regard, the information collection unit 251 acquires identification information of the user terminal 100 receiving the advertisement executed by the advertisement execution unit 255, and based on the identification information of the user terminal 100, the user terminal ( The user behavior log including the category of the advertisement provided in the advertisement, whether the advertisement is selected, and whether to purchase a product based on the advertisement, etc. may be collected as the user data 210 and stored and managed in the memory 230. The information collecting unit 251 may collect user data 210 whenever an advertisement is provided to the user terminals 100. The information collecting unit 251 may collect the user data 210 for a specified period or collect a specified amount, and if the specified condition is satisfied, transmit the user data 210 to the parameter extraction unit 253. For example, the information collection unit 251 may transmit the collected user data 210 to the parameter extraction unit 253 when the specified period has elapsed or the user data 210 is collected in a specified amount. Alternatively, the information collection unit 251 may transmit the user data 210 to the parameter extraction unit 253 to extract parameters corresponding to the corresponding category when the advertisement execution unit 255 wants to execute advertisements of a specific category. have.

In the step of analyzing the user data 210, the parameter extraction unit 253 may load the user data 210 and reduce the analysis amount through the pre-processing process for the loaded user data. For example, a user who selects an advertisement related to item A or a user who purchases a product through advertisement related to item A has one person, and the user selects advertisement related to item B, or an advertisement related to item B. If there is only one second user who purchased the product through, the similarity between item A and B is 0. That is, the behavior log of a person who purchases only one item among all data may be insignificant data in similarity calculation. After removing the data, the parameter extraction unit 253 may secure clean data as shown in Table 1.

user Category_1 Category_2 Category_3 Category_4 Category_5 One One 0 0 0 0 2 0 0 0 One 0 3 One 0 0 0 0 4 0 0 0 One 0 5 0 One One 0 0 6 0 One 0 One One

In Table 1, a case in which a user has a history of selecting an advertisement for each category or purchasing a product through advertisement may be 1. When the user data classification value as shown in Table 1 is obtained, the parameter extraction unit 253 may obtain a table as shown in Table 2 using at least one of various logics included in the recommendation logic 230. Here, Table 2 is a value obtained by using jaccard similarity.

user Category_1 Category_2 Category_3 Category_4 Category_5 Category_1 One Category_2 0.78 One Category_3 0.21 0.23 One Category_4 0.14 0.2 0.456 One Category_5 0.99 0.1 0.121 0.111 One

Table 2 above shows the similarity between categories in the form of a table, as mentioned above. Meanwhile, the value of User Score determines the priority of the preference between the user and the corresponding category (advertising material). Can be In relation to a method for obtaining a user score value, for example, a method for obtaining user scores for user 1 and category 1, the parameter extraction unit 253 first obtains the category most similar to category 1 through Table 2. That is, the parameter extraction unit 253 obtains the similarity result values in the order of categories 5, 2, 3, and 4. The parameter extraction unit 253 puts the obtained similarity value into the similarities value of Equation 1 below.

[Equation 1]

User score = sum (history * similarities) / sum (similarities)

The history value may be a value (eg, 1 or 0) indicating whether the user has purchased the corresponding product in Table 1. The user score according to Equation 1 may be set to a value obtained by multiplying the usage history value for each advertisement of a category by multiplying the similarity scores between the categories and the sum of the similarity scores between the categories. Alternatively, a value obtained by multiplying the similarity scores between the categories by the usage history value for each advertisement of a category is a molecular value, and the user score is a value that is the sum of the similarity scores between the categories as a denominator value. Can be.

For example, after multiplying 0.99, 0.78, 0.21.0.14 and (0,0,0,0), the sum of the sums becomes a numerator, and only the similarity value added becomes the denominator, so the user score for category 1 of user 1 is 0. Can be. As described above, if user 2 proceeds to Category 2, 3, 4, and 5, and then continues to users 2 to 6, the final user score can be obtained as shown in Table 3.

user Category_1 Category_2 Category_3 Category_4 Category_5 One 0 0.234 0.266 0.1181 0 2 0.082343 0 0.095912 0.088885 3 0 0.089767 0 0 0 4 0.082343 0.083568 0 0.083502 0 5 0 0 0.114305 0.093197 0.088221 6 0 0 0 0 0

In Table 3, when ranking by category for each user, values as shown in Table 4 may be obtained.

user 1st rank 2nd rank 3 rank 4th rank 5th rank One Category_3 Category_2 Category_4 2 Category_3 Category_5 Category_1 3 Category_2 4 Category_2 Category_4 Category_1 5 Category_3 Category_4 Category_5 6

The values shown in Table 4 may be used as user recommendation information. If group_by the result value shown in Table 4 by category is obtained, Table 5 and the final result can be obtained.

1st priority parameter 2nd priority parameter 3rd priority parameter Category_1 2/4 Category_2 3/4 One Category_3 1/2/5 Category_4 4/5 Category_5 5 Category_6

Table 5 shows the parameters for each category. As shown in Table 5, if the first priority parameter is used when the advertisement is first sent and the parameter is insufficient afterwards, the second priority parameter can be utilized to lookalike the parameter. In addition, since the above tables can be calculated daily or hourly, the freshness of the parameters is always maintained.

The advertisement executive unit 255 may transmit advertisements for advertisements for each category using Table 5. For example, the advertisement execution unit 255 may transmit advertisements to the user terminal 2 and the user terminal 4 when the category 1 advertisement is to be transmitted. In addition, the advertisement execution unit 255, if you want to send the advertisement of category 2, first performs the advertisement transmission to the user terminal 3 and the user terminal 4, and if additional advertisement transmission is required, the advertisement transmission to the user terminal 1 can do. As described above, the advertisement execution unit 255 may perform advertisement transmission according to parameters for each category, thereby providing advertisements of categories of interest to the user of the user terminal according to the behavior log of the user terminal, which is useful for the user. Advertising can be provided. As a result, the advertisement provided by the advertisement executive unit 255 is selected by the user, and it can be induced to easily purchase products through the advertisement.

6 is a diagram illustrating an example of a method of operating a server device related to advertisement operation according to an embodiment of the present invention.

Referring to FIG. 6, in the method of operating a server device related to the advertisement operation of the present invention, in step 601, the processor 250 of the server device 200 may collect user data. In this regard, the processor 250 may provide advertisements of a specific category to a specific user terminal 100, and collect a behavior log for the advertisements from the user terminal 100 as user data. For example, the processor 250 may provide advertisement identification information capable of identifying advertisements provided from the user terminal 100, advertisement selection or playback information, product purchase history information based on advertisements, and the like, with respect to provided advertisements. Can be collected. The processor 250 may provide at least one advertisement to each of the plurality of user terminals 100, and collect action logs for advertisements provided from each of the user terminals 100.

In step 603, the processor 250 may check whether a designated time has arrived. The designated time may include a preset period to extract parameters based on user data. Alternatively, the designated time may be a time when the number of user data is acquired over a specified number. When the designated time has not arrived, the processor 250 may branch to step 601 and perform the following operation again. For example, the processor 250 may collect action logs for advertisements from at least one user terminal 100.

When the designated time has arrived, the processor 250 may perform parameter extraction in step 605. For example, the processor 250 may perform filtering on user data to remove unimportant data in the similarity calculation. For example, the processor 250 may remove user data having a similarity score of zero. The processor 250 may apply the recommendation logic specified in the user data to calculate the similarity score for each category of the advertisement. The processor 250 may calculate a user score for each user terminal based on the calculated similarity score for each category. The processor 250 may calculate a ranking value for each user category based on the calculated user scores for each user terminal. The processor 250 may calculate a parameter for each category by grouping the calculated ranking values for each user (or for each user terminal) by category. The processor 250 may continuously perform the above-described parameter calculation operation for each category in a designated time unit.

In step 607, the processor 250 may perform advertisement transmission based on the extracted parameter. For example, the processor 250 may determine which category advertisement is provided to which user terminal based on the extracted parameter. The processor 250 may transmit a designated category to a specific user terminal according to the determination.

In step 609, the processor 250 may check whether an event for ending the advertisement sending operation occurs. For example, the processor 250 may determine that an end event related to an advertisement sending operation occurs when a specified time has passed for the advertisement sending. Alternatively, when the user action log for selecting a sent advertisement or a user action log for purchasing a specific product based on the sent advertisement occurs, the processor 250 may determine that an end event of the advertisement transmission operation is generated.

If there is no termination event related to the advertisement sending operation, the processor 250 may branch to step 601 and perform the following operation again. When an end event occurs, the processor 250 may end collecting user data and maintain a standby state for connection with the user terminal 100.

As described above, according to the server device supporting the calculation of parameters for the advertisement transmission according to an embodiment of the present invention and the advertisement transmission using the parameters calculated thereon, and the operation method thereof, related to the advertisement transmission in a logical manner By calculating the parameters, it is possible to provide advertisements of a more effective category to user terminals.

The present invention is applied to a field related to advertisement transmission, and particularly, to a method and a server device related to parameter output to be applied to advertisement transmission.

The present invention can logically calculate the user score based on the similarity score for each category of the advertisement, and prioritize the transmission of advertisements based on the user score, so that more useful advertisements can be appropriately provided to users in need. Support.

10: Advertising operation system
50: network
200: server device
201: Data collection server
203: recommended engine server
205: ad server
210: communication circuit
230: memory
250: processor
251: information collection department
253: parameter extraction unit
255: advertising executive

Claims (11)

Communication circuit;
A memory for storing user data related to use of advertisements for at least one category collected from at least one user terminal connected through the communication circuit;
And a processor functionally connected to the communication circuit and the memory.
The processor
Similarity scores between categories of a plurality of advertisements are calculated based on the user data, and user scores for categories for each user terminal using at least one advertisement among the plurality of advertisements are calculated based on the similarity scores. A server device supporting parameter extraction, characterized in that it is configured to calculate parameters for each category based on user scores and to store the calculated parameters for each category. According to claim 1,
The processor
A server device for supporting parameter extraction, characterized in that it is configured to determine the ranking of categories for each user terminal using the user scores and to calculate the parameters for each category using the ranking of the categories for each user terminal. According to claim 1,
The processor
A server device supporting parameter extraction, characterized in that it is configured to filter and remove data having a similarity of 0 among the data included in the user data. According to claim 1,
The processor
A server device supporting parameter extraction, characterized in that it is set to calculate similarity scores among a plurality of advertisement categories of the user data using a cosine similarity algorithm or a jacquard similarity algorithm. According to claim 1,
The processor
Regarding a specific user terminal, the value obtained by multiplying the similarity scores between the categories by the usage history values of advertisements of a plurality of categories is a molecular value, and the value of the sum of the similarity scores between the categories is a denominator value. A server device supporting parameter extraction characterized in that it is set to a user score of a specific user terminal. According to claim 1,
The processor
A server device supporting parameter extraction, characterized in that the index values of user terminals, which are prioritized for each of the plurality of advertisement categories, are set as parameters for each category. According to claim 1,
The processor
A server device supporting parameter extraction, characterized in that it is configured to determine a specific user terminal that transmits an advertisement of a specific category by checking the parameters for each category. The method of claim 7,
The processor
If there is no highest priority parameter of a specific category, the server device supporting parameter extraction, characterized in that it is set to send the advertisement of the specific category based on the next highest priority parameter. Server device,
Collecting user data related to the use of at least one category advertisement from at least one user terminal connected through a communication circuit;
Calculating similarity scores between categories of a plurality of advertisements based on the user data;
Calculating user scores for categories for each user terminal using at least one advertisement among the plurality of advertisements based on the similarity scores;
And calculating parameters for each category based on the user scores. The method of claim 9,
Determining a ranking of categories for each user terminal using the user scores;
Filtering and removing data having a similarity of 0 among the data included in the user data; or
Checking the parameters for each category and sending an advertisement of a specific category to a specific user terminal; Method of extracting parameters of a server device, characterized in that it further comprises at least one step. The method of claim 9,
The step of calculating the user score is
Regarding a specific user terminal, the value obtained by multiplying the similarity scores between the categories by the usage history values of advertisements of a plurality of categories is a molecular value, and the value of the sum of the similarity scores between the categories is a denominator value. And calculating the user score of a specific user terminal.

Images