CN112449217B - Method and device for pushing video, electronic equipment and computer readable medium - Google Patents
Method and device for pushing video, electronic equipment and computer readable medium Download PDFInfo
- Publication number
- CN112449217B CN112449217B CN201910822637.1A CN201910822637A CN112449217B CN 112449217 B CN112449217 B CN 112449217B CN 201910822637 A CN201910822637 A CN 201910822637A CN 112449217 B CN112449217 B CN 112449217B
- Authority
- CN
- China
- Prior art keywords
- video
- user
- target
- message
- characteristic data
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 40
- 238000012549 training Methods 0.000 claims abstract description 54
- 238000013528 artificial neural network Methods 0.000 claims abstract description 18
- 238000004891 communication Methods 0.000 claims description 26
- 238000012216 screening Methods 0.000 claims description 12
- 239000013598 vector Substances 0.000 claims description 10
- 238000004590 computer program Methods 0.000 claims description 9
- 238000010586 diagram Methods 0.000 description 16
- 230000006870 function Effects 0.000 description 10
- 230000008569 process Effects 0.000 description 6
- 238000011176 pooling Methods 0.000 description 5
- 238000012545 processing Methods 0.000 description 5
- 230000004913 activation Effects 0.000 description 4
- 230000003287 optical effect Effects 0.000 description 4
- 230000008901 benefit Effects 0.000 description 3
- 230000000694 effects Effects 0.000 description 3
- 239000011159 matrix material Substances 0.000 description 3
- 230000004048 modification Effects 0.000 description 3
- 238000012986 modification Methods 0.000 description 3
- 230000006399 behavior Effects 0.000 description 2
- 230000005540 biological transmission Effects 0.000 description 2
- 238000013461 design Methods 0.000 description 2
- 239000000835 fiber Substances 0.000 description 2
- 230000005764 inhibitory process Effects 0.000 description 2
- 230000003993 interaction Effects 0.000 description 2
- 230000000670 limiting effect Effects 0.000 description 2
- 210000002569 neuron Anatomy 0.000 description 2
- 230000000644 propagated effect Effects 0.000 description 2
- 230000002829 reductive effect Effects 0.000 description 2
- 239000004065 semiconductor Substances 0.000 description 2
- 238000004458 analytical method Methods 0.000 description 1
- 230000002238 attenuated effect Effects 0.000 description 1
- 238000004422 calculation algorithm Methods 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 238000013527 convolutional neural network Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000008034 disappearance Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 230000005284 excitation Effects 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 239000004973 liquid crystal related substance Substances 0.000 description 1
- 239000013307 optical fiber Substances 0.000 description 1
- 230000036961 partial effect Effects 0.000 description 1
- 238000003672 processing method Methods 0.000 description 1
- 230000002441 reversible effect Effects 0.000 description 1
- 230000003997 social interaction Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Images
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
- H04N21/20—Servers specifically adapted for the distribution of content, e.g. VOD servers; Operations thereof
- H04N21/25—Management operations performed by the server for facilitating the content distribution or administrating data related to end-users or client devices, e.g. end-user or client device authentication, learning user preferences for recommending movies
- H04N21/258—Client or end-user data management, e.g. managing client capabilities, user preferences or demographics, processing of multiple end-users preferences to derive collaborative data
- H04N21/25866—Management of end-user data
- H04N21/25891—Management of end-user data being end-user preferences
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
- H04N21/20—Servers specifically adapted for the distribution of content, e.g. VOD servers; Operations thereof
- H04N21/25—Management operations performed by the server for facilitating the content distribution or administrating data related to end-users or client devices, e.g. end-user or client device authentication, learning user preferences for recommending movies
- H04N21/251—Learning process for intelligent management, e.g. learning user preferences for recommending movies
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
- H04N21/20—Servers specifically adapted for the distribution of content, e.g. VOD servers; Operations thereof
- H04N21/25—Management operations performed by the server for facilitating the content distribution or administrating data related to end-users or client devices, e.g. end-user or client device authentication, learning user preferences for recommending movies
- H04N21/251—Learning process for intelligent management, e.g. learning user preferences for recommending movies
- H04N21/252—Processing of multiple end-users' preferences to derive collaborative data
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
- H04N21/40—Client devices specifically adapted for the reception of or interaction with content, e.g. set-top-box [STB]; Operations thereof
- H04N21/43—Processing of content or additional data, e.g. demultiplexing additional data from a digital video stream; Elementary client operations, e.g. monitoring of home network or synchronising decoder's clock; Client middleware
- H04N21/442—Monitoring of processes or resources, e.g. detecting the failure of a recording device, monitoring the downstream bandwidth, the number of times a movie has been viewed, the storage space available from the internal hard disk
- H04N21/44213—Monitoring of end-user related data
- H04N21/44222—Analytics of user selections, e.g. selection of programs or purchase activity
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
- H04N21/40—Client devices specifically adapted for the reception of or interaction with content, e.g. set-top-box [STB]; Operations thereof
- H04N21/45—Management operations performed by the client for facilitating the reception of or the interaction with the content or administrating data related to the end-user or to the client device itself, e.g. learning user preferences for recommending movies, resolving scheduling conflicts
- H04N21/4508—Management of client data or end-user data
- H04N21/4532—Management of client data or end-user data involving end-user characteristics, e.g. viewer profile, preferences
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
- H04N21/40—Client devices specifically adapted for the reception of or interaction with content, e.g. set-top-box [STB]; Operations thereof
- H04N21/45—Management operations performed by the client for facilitating the reception of or the interaction with the content or administrating data related to the end-user or to the client device itself, e.g. learning user preferences for recommending movies, resolving scheduling conflicts
- H04N21/466—Learning process for intelligent management, e.g. learning user preferences for recommending movies
- H04N21/4662—Learning process for intelligent management, e.g. learning user preferences for recommending movies characterized by learning algorithms
- H04N21/4666—Learning process for intelligent management, e.g. learning user preferences for recommending movies characterized by learning algorithms using neural networks, e.g. processing the feedback provided by the user
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
- H04N21/40—Client devices specifically adapted for the reception of or interaction with content, e.g. set-top-box [STB]; Operations thereof
- H04N21/45—Management operations performed by the client for facilitating the reception of or the interaction with the content or administrating data related to the end-user or to the client device itself, e.g. learning user preferences for recommending movies, resolving scheduling conflicts
- H04N21/466—Learning process for intelligent management, e.g. learning user preferences for recommending movies
- H04N21/4667—Processing of monitored end-user data, e.g. trend analysis based on the log file of viewer selections
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
- H04N21/40—Client devices specifically adapted for the reception of or interaction with content, e.g. set-top-box [STB]; Operations thereof
- H04N21/45—Management operations performed by the client for facilitating the reception of or the interaction with the content or administrating data related to the end-user or to the client device itself, e.g. learning user preferences for recommending movies, resolving scheduling conflicts
- H04N21/466—Learning process for intelligent management, e.g. learning user preferences for recommending movies
- H04N21/4668—Learning process for intelligent management, e.g. learning user preferences for recommending movies for recommending content, e.g. movies
Landscapes
- Engineering & Computer Science (AREA)
- Databases & Information Systems (AREA)
- Signal Processing (AREA)
- Multimedia (AREA)
- Social Psychology (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Computing Systems (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Computation (AREA)
- Computer Networks & Wireless Communication (AREA)
- Computer Graphics (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
- Information Transfer Between Computers (AREA)
Abstract
The invention discloses a method and a device for pushing videos, and relates to the technical field of computers. One embodiment of the method comprises: determining a user group corresponding to a target video according to the target video clicked by a target user; determining a video to be recommended according to the characteristic data of the user group and a video recommendation model; pushing a message to the target user; the message carries an identifier of the video to be recommended; the video recommendation model is obtained by training a neural network based on training characteristic data of a video and corresponding label data. The implementation mode can solve the technical problems that the pushed video content is single and no distinction is made.
Description
Technical Field
The invention relates to the technical field of computers, in particular to a method and a device for pushing videos.
Background
The short video is a short video, is an internet content transmission mode, and is generally a video transmitted on an internet new medium within 5 minutes; with the popularization of mobile terminals and the increasing speed of networks, short and fast mass flow transmission contents are gradually favored by various large platforms, fans and capital. Therefore, with the development of the internet and big data, it is becoming better and better for users to enjoy leisure and entertainment.
In the process of implementing the invention, the inventor finds that at least the following problems exist in the prior art:
the conventional short video message center pushes the videos of the same type recently published or browsed by a user attention person, the pushed video content is single, and the user experience is poor due to no distinction degree.
Disclosure of Invention
In view of this, embodiments of the present invention provide a method and an apparatus for pushing a video, so as to solve the technical problems that the content of the pushed video is relatively single and there is no distinction.
To achieve the above object, according to an aspect of the embodiments of the present invention, there is provided a method for pushing a video, including:
determining a user group corresponding to a target video according to the target video clicked by a target user;
determining a video to be recommended according to the characteristic data of the user group and a video recommendation model;
pushing a message to the target user;
the message carries the identification of the video to be recommended, and the video recommendation model is obtained by training a neural network based on training characteristic data of the video and corresponding label data.
Optionally, determining a video to be recommended according to the feature data of the user group and a video recommendation model includes:
determining the characteristic data of the user group according to the user characteristic data of the user group;
inputting the characteristic data of the user group into a video recommendation model so as to output at least one video identifier and a corresponding preset operation probability;
and according to the preset operation probability, performing descending order arrangement on the at least one video, and screening out a preset number of videos as videos to be recommended.
Optionally, determining, according to a target video clicked by a target user, a user group corresponding to the target video, including:
according to a target video clicked by a target user, taking each user paying attention to the target video as a user group corresponding to the target video; or,
determining a publishing user publishing the target video according to the target video clicked by the target user, and taking each user concerning the publishing user as a user group corresponding to the target video.
Optionally, the training feature data of the video includes: the characteristic data of the video, the characteristic data of the article corresponding to the video, and the characteristic data of the user who performs preset operation on the article;
the tag data of the video includes: whether to execute preset operation on the article corresponding to the video and/or whether to execute preset operation on the video.
Optionally, pushing a message to the target user includes:
determining a message pushing mode of the target user according to a message pushing model; the message pushing model is obtained by training a neural network based on message characteristic data and corresponding label data;
and pushing the message to the target user by adopting the message pushing mode.
Optionally, after the message is pushed to the target user, the method further includes:
constructing a feature vector corresponding to each user according to the feature data of each user in the user group, the feature data of the video browsed by each user and the time sequence feature of the video browsed by each user;
and respectively calculating the similarity among the feature vectors, screening a preset number of communication users from the user group based on the sequence of the similarity from large to small, and establishing a communication association relationship among the communication users.
In addition, according to another aspect of the embodiments of the present invention, there is provided an apparatus for pushing a video, including:
the user group module is used for determining a user group corresponding to a target video according to the target video clicked by a target user;
the determining module is used for determining a video to be recommended according to the characteristic data of the user group and the video recommendation model;
the pushing module is used for pushing a message to the target user;
the message carries an identifier of the video to be recommended; the video recommendation model is obtained by training a neural network based on training characteristic data of a video and corresponding label data.
Optionally, the determining module is further configured to:
determining the characteristic data of the user group according to the user characteristic data of the user group;
inputting the characteristic data of the user group into a video recommendation model so as to output at least one video identifier and a corresponding preset operation probability;
and according to the preset operation probability, performing descending order on the at least one video, and screening a preset number of videos from the videos as videos to be recommended.
Optionally, the user group module is further configured to:
according to a target video clicked by a target user, taking each user concerning the target video as a user group corresponding to the target video; or,
determining a publishing user publishing the target video according to the target video clicked by the target user, and taking each user concerning the publishing user as a user group corresponding to the target video.
Optionally, the training feature data of the video includes: the characteristic data of the video, the characteristic data of the article corresponding to the video, and the characteristic data of the user who performs preset operation on the article;
the tag data of the video includes: whether to execute preset operation on the article corresponding to the video and/or whether to execute preset operation on the video.
Optionally, the pushing module is further configured to:
determining a message pushing mode of the target user according to a message pushing model; the message pushing model is obtained by training a neural network based on message characteristic data and corresponding label data;
and pushing the message to the target user by adopting the message pushing mode.
Optionally, the pushing module is further configured to: after the message is pushed to the target user, constructing a feature vector corresponding to each user according to the feature data of each user in the user group, the feature data of the video browsed by each user and the time sequence feature of the video browsed by each user;
and respectively calculating the similarity among the feature vectors, screening a preset number of communication users from the user group based on the sequence of the similarity from large to small, and establishing a communication association relation among the communication users.
According to another aspect of the embodiments of the present invention, there is also provided an electronic device, including:
one or more processors;
a storage device for storing one or more programs,
when executed by the one or more processors, cause the one or more processors to implement the method of any of the embodiments described above.
According to another aspect of the embodiments of the present invention, there is also provided a computer readable medium, on which a computer program is stored, which when executed by a processor implements the method of any of the above embodiments.
One embodiment of the above invention has the following advantages or benefits: the technical means that the user group corresponding to the target video is determined according to the target video clicked by the target user, and the video to be recommended is determined through the characteristic data of the user group and the video recommendation model is adopted, so that the technical problems that the pushed video content is single and no distinction degree exists in the prior art are solved. The embodiment of the invention determines the video which is interested by the user by utilizing the characteristic data of the user group, is easier to attract the attention of the user, improves the click rate of the video, can enrich the content of the pushed video, avoids pushing a single video to the user, and therefore, the user requirement is better grasped.
Further effects of the above-mentioned non-conventional alternatives will be described below in connection with the embodiments.
Drawings
The drawings are included to provide a better understanding of the invention and are not to be construed as unduly limiting the invention. Wherein:
fig. 1 is a schematic diagram of a main flow of a method of pushing video according to an embodiment of the present invention;
fig. 2 is a schematic diagram of a main flow of a method for pushing video according to one referential embodiment of the present invention;
FIG. 3 is a schematic diagram of interaction between a client and a server according to an embodiment of the present invention;
FIG. 4 is a schematic diagram of the main blocks of an apparatus for pushing video according to an embodiment of the present invention;
FIG. 5 is an exemplary system architecture diagram in which embodiments of the present invention may be employed;
fig. 6 is a schematic block diagram of a computer system suitable for use in implementing a terminal device or server of an embodiment of the invention.
Detailed Description
Exemplary embodiments of the present invention are described below with reference to the accompanying drawings, in which various details of embodiments of the invention are included to assist understanding, and which are to be considered as merely exemplary. Accordingly, those of ordinary skill in the art will recognize that various changes and modifications of the embodiments described herein can be made without departing from the scope and spirit of the invention. Also, descriptions of well-known functions and constructions are omitted in the following description for clarity and conciseness.
Fig. 1 is a schematic diagram of a main flow of a method of pushing a video according to an embodiment of the present invention. As an embodiment of the present invention, as shown in fig. 1, the method for pushing a video may include:
step 101, determining a user group corresponding to a target video according to the target video clicked by a target user.
In this step, a user group corresponding to the target video may be determined according to the target video clicked by the target user, so that the determination is performed based on the feature data in the user group in step 102.
Optionally, step 101 comprises: according to a target video clicked by a target user, taking each user concerning the target video as a user group corresponding to the target video; or determining a publishing user publishing the target video according to the target video clicked by the target user, and taking each user concerning the publishing user as a user group corresponding to the target video. For example, if 1000 users who focus on a certain video are present, the 1000 users may be regarded as a user group; alternatively, if the user who pays attention to a distribution user of a certain video is 20000 people, the 20000 users can also be a user group.
And 102, determining a video to be recommended according to the characteristic data of the user group and the video recommendation model.
The video recommendation model is obtained by training a neural network based on training feature data of a video and corresponding label data. Therefore, before step 102, the video recommendation model needs to be trained.
Optionally, the neural network is trained based on training feature data of the video and corresponding label data thereof to obtain a video recommendation model through training. Specifically, firstly, training samples, namely training feature data of the video and corresponding label data, need to be constructed in advance, then the training samples are input into a neural network, and a video recommendation model is obtained through multiple iterative training.
Optionally, the training feature data of the video includes: the characteristic data of the video, the characteristic data of the article corresponding to the video, and the characteristic data of the user who performs preset operation on the article. For example, the feature data of the video may include the number of clicked videos, the number of shared videos, the number of praised videos, the number of comments, and the like, and these feature data are generally displayed in a video introduction page; the brand, category and the like of the article corresponding to the video can be included. For example, the characteristic data of the item may include price, color, size, season information, etc. of the item, and may also include the number of clicks on the item over time, the number of orders placed, the number of shopping carts entered, the number of concerns. For another example, the feature data of the user may include a user gender, a user age, a user region, whether the user is a high-level member, a user activity, a user preference, information of a user terminal device, and the like.
It should be noted that one video may correspond to a plurality of articles and be used as a plurality of training samples, that is, one video is used as a training sample every time it corresponds to an article. The number of users performing the preset operation on the article may also be multiple, and 100 training samples may be obtained assuming that 100 users perform the preset operation on a certain article.
Optionally, the tag data of the video includes: whether to execute preset operation on the article corresponding to the video and/or whether to execute preset operation on the video. The preset operation can be clicking, ordering, sharing, commenting, praising, paying attention to and the like. If preset operation is executed on the video or the article corresponding to the video, marking the training sample as a positive sample; and if the preset operation is not executed on the video or the article corresponding to the video, marking the training sample as a negative sample.
Therefore, after the training sample is constructed, the training feature data of the video and the corresponding label data can be adopted, and the video recommendation model can be obtained by training through an error back propagation algorithm. Alternatively, the neural network may be a convolutional neural network, typically comprising several convolutional layers, an activation layer, a pooling layer, and a fully-connected layer.
There are three more important parameters for the convolutional layer: the method comprises the steps of convolution kernel depth, stride and zero padding, wherein the depth of the convolution kernels corresponds to the number of the convolution kernels, each convolution kernel can only extract partial features of input data, and the extraction of features by a single convolution kernel is insufficient; the step refers to the number of pixel units of a sliding filter matrix on an input matrix, and the larger the step is, the smaller the obtained characteristic diagram is; in many scenarios, the size of the convolution kernel is not necessarily evenly divided by the dimension of the input data matrix, and there are two processing methods: valid padding and same padding.
An active layer: the variety of the activation function is more, relu is selected as the activation function, and compared with a sigmoid function, relu has the advantages that: unilateral inhibition, when the input is less than 0, the neuron is in an inhibition state, and when the input is more than 0, the neuron is in an activation state, and the convergence rate is high; a relatively broad excitation boundary; when the input is larger than 0, the gradient is kept not to be attenuated, so that the problem of gradient disappearance is relieved.
A pooling layer: pooling is the process of integrating local features to obtain new features, and the purpose of pooling layer design is two: firstly, the data size of the characteristics of the next layer of samples to be processed is reduced, secondly, the parameter quantity is reduced, overfitting is prevented, and the method adopts a maximum pooling mode.
Full connection layer: the fully connected layer is a distributed feature representation pre-learned by the previous layers, is mapped to a sample mark space, and then a loss function is utilized to regulate and control the learning process.
After the video recommendation model is obtained through training, the feature data of the user group obtained in step 101 is input into the video recommendation model to determine a video to be recommended. Optionally, determining a video to be recommended according to the feature data of the user group and a video recommendation model includes: determining the characteristic data of the user group according to the user characteristic data of the user group; inputting the characteristic data of the user group into the video recommendation model to output at least one video identifier and a corresponding preset operation probability; and according to the preset operation probability, performing descending order on the at least one video, and screening a preset number of videos from the videos as videos to be recommended. The feature data of the user group may be gender, region, preferred categories, preferred stores, purchasing power, interested items, interested videos, and the like of most users in the user group, and may be obtained by analyzing the user feature data of each user in the user group.
Step 103, pushing a message to the target user.
Since the video to be recommended is determined through step 102, a message may be pushed to a target user, where the message carries an identifier of the video to be recommended.
For each user group, the ages, hobbies and the like of the users in the group are different, so that the user requirements are greatly different. Optionally, the method may further include: constructing a feature vector corresponding to each user according to the feature data of each user in the user group, the feature data of the video browsed by each user and the time sequence feature of the video browsed by each user; and respectively calculating the similarity among the feature vectors, screening a preset number of communication users from the user group based on the sequence of the similarity from large to small, and establishing a communication association relation among the communication users.
For example, if the similarity between the user a and the user B in a certain user group is high, a communication association relationship may be established between the user a and the user B, so as to facilitate communication and intercommunication between the user a and the user B, for example, sending information to each other. Specifically, a communication association relationship can be established between the user a and the user B through a network connection technology, so that the user a and the user B can send information to each other, and contact is facilitated. Of course, a communication association relationship may also be established between more users, which is not limited in this embodiment of the present invention.
According to the various embodiments, the technical means that the user group corresponding to the target video is determined according to the target video clicked by the target user, so that the video to be recommended is determined through the characteristic data of the user group and the video recommendation model can be seen, and the technical problems that the pushed video content is single and no distinction exists in the prior art are solved. The embodiment of the invention determines the video which is interested by the user by utilizing the characteristic data of the user group, is easier to attract the attention of the user, improves the click rate of the video, can enrich the content of the pushed video, avoids pushing a single video to the user, and therefore, the user requirement is better grasped.
Fig. 2 is a schematic diagram of a main flow of a method for pushing video according to one referential embodiment of the present invention.
Step 201, training a neural network based on training feature data of the video and corresponding label data thereof to obtain a video recommendation model through training.
Firstly, a training sample, namely training characteristic data of a video and corresponding label data, needs to be constructed in advance, then the training sample is input into a neural network, and a video recommendation model is obtained through multiple iterative training. Optionally, the training feature data of the video includes: the characteristic data of the video, the characteristic data of the article corresponding to the video, and the characteristic data of the user who performs preset operation on the article.
Step 202, training a neural network based on the message characteristic data and the corresponding label data thereof to obtain a message pushing model through training.
Optionally, the message characteristic data may include message push time, message file, message category, etc., message push modality, push user, number of message clicks, etc.
For example, the message pushing time can be roughly divided into 9-10 am, 12-14 am, 17-20 am, 21-22 am; the message pattern may include title, content, picture, sound, etc., and the message categories may include a time greeting type, an inclusive culture style type, a plan to try to get close to a user type, and a wild type; the message push modality may include a popup and a message box.
Specifically, the message characteristic data can be derived from a message sending table and a message click table, and a message id, message pushing time, message category and pushing user can be obtained from the message sending table; the number of times of message clicks and other fine-grained indexes of the message can be obtained from the message click table. The message characteristic data is then processed, which may include the steps of: acquiring a user message sending record of the last month from a message sending table; acquiring a user message click record of the latest month from a message click table; correlating the record tables obtained in the first step and the second step to generate a user message sending click record broad table, thereby obtaining the sending message id, the message content, the sending time, whether to click, the click time and the environmental information (equipment number, geographical position and the like) when clicking for each user; forming a message sensitive user pool by all users who click messages in the last month, and acquiring user image information based on click list data, such as: the information of the user information comprises (a) user region information, (b) user mobile phone equipment information (mobile phone terminal, mobile phone brand, network environment), (c) user message sending history, (d) user message behavior activity, (e) user message behavior span (message click days, click weeks), (f) message time preference clicked by the user, region preference, purchasing power preference, sex age preference, (g) user average message click rate (average click rate per hour), and the like.
Optionally, the tag data corresponding to the message feature data includes: whether to execute preset operation on the article corresponding to the message and/or whether to execute preset operation on the message. The preset operation can be clicking, ordering, sharing, commenting, praising, paying attention to and the like. If preset operation is executed on the message or the article corresponding to the message, marking the training sample as a positive sample; and if the preset operation is not executed on the message or the article corresponding to the message, marking the training sample as a negative sample.
Step 203, determining a user group corresponding to a target video according to the target video clicked by the target user.
Specifically, according to a target video clicked by a target user, each user who pays attention to the target video can be used as a user group corresponding to the target video; or, according to a target video clicked by a target user, a publishing user who publishes the target video may be determined, and each user who concerns the publishing user is taken as a user group corresponding to the target video.
Step 204, determining the characteristic data of the user group according to the user characteristic data of the user group.
The feature data of the user group may be gender, region, preferred categories, preferred stores, purchasing power, interested items, interested videos, and the like of most users in the user group, and may be obtained by analyzing the user feature data of each user in the user group.
The video recommendation model can output each video label and the corresponding preset operation probability thereof based on the input feature data of the user group. The output quantity of the video recommendation model can be preset, so that the preset quantity of video identifications and the corresponding preset operation probability can be output.
And step 206, performing descending order on the at least one video according to the preset operation probability, and screening a preset number of videos from the videos as videos to be recommended.
The closer the ranking is, the more interesting the target user is in the video to be recommended, and the click rate of the video is favorably improved.
Step 207, determining the message pushing mode of the target user according to the message pushing model, and pushing a message to the target user by adopting the message pushing mode.
The user characteristic data of the target user is input into the message pushing model, and the message pushing mode, such as message pushing time, message pattern, message type and the like, and the message pushing form can be determined, so that deep personalization of message pushing is realized, user experience is improved, and the click rate of the user is increased.
In addition, in a reference embodiment of the present invention, the implementation content of the method for pushing video is described in detail in the above-mentioned method for pushing video, so that the repeated content is not described herein.
Fig. 3 is a schematic diagram of interaction between a client and a server according to an embodiment of the present invention. As shown in fig. 3, the first step: a user logs in a short video App; the second step is that: a user clicks a message center of a short video App; the third step: and after the user enters the message center, starting social interaction. The server side comprises a service server cluster, a message queue, a calculation server cluster and a database. When a client logs in, a service request is sent, a service cluster server collects characteristic data and distributes tasks, request results can be synchronously returned to the client, resource operations (such as model training, result determination and the like) are requested to be executed from a computing server cluster, the requests can be firstly pushed to a message queue, and the computing server cluster acquires the requests from the message queue and executes corresponding tasks. And the computing server cluster pushes the processing result to a message queue, the service server cluster receives the message and analyzes the message source, and the analysis result is asynchronously pushed to the client. The database is used for storing the parameters of the video push model and the parameters of the message push model.
Fig. 4 is a schematic diagram of main modules of an apparatus for pushing video according to an embodiment of the present invention, and as shown in fig. 4, the apparatus for pushing video 400 includes a user group module 401, a determination module 402, and a pushing module 403. The user group module 401 is configured to determine, according to a target video clicked by a target user, a user group corresponding to the target video; the determining module 402 is configured to determine a video to be recommended according to the feature data of the user group and a video recommendation model; the pushing module 403 is configured to push a message to the target user. The message carries an identifier of the video to be recommended; the video recommendation model is obtained by training a neural network based on training feature data of a video and corresponding label data.
Optionally, the determining module 402 is further configured to:
determining the characteristic data of the user group according to the user characteristic data of the user group;
inputting the characteristic data of the user group into a video recommendation model so as to output at least one video identifier and a corresponding preset operation probability;
and according to the preset operation probability, performing descending order arrangement on the at least one video, and screening out a preset number of videos as videos to be recommended.
Optionally, the user group module 401 is further configured to:
according to a target video clicked by a target user, taking each user paying attention to the target video as a user group corresponding to the target video; or,
determining a publishing user publishing the target video according to the target video clicked by the target user, and taking each user concerning the publishing user as a user group corresponding to the target video.
Optionally, the training feature data of the video includes: the characteristic data of the video, the characteristic data of the article corresponding to the video, and the characteristic data of the user who performs preset operation on the article;
the tag data of the video includes: whether to execute preset operation on the article corresponding to the video and/or whether to execute preset operation on the video.
Optionally, the pushing module 403 is further configured to:
determining a message pushing mode of the target user according to a message pushing model; the message pushing model is obtained by training a neural network based on message characteristic data and corresponding label data;
and pushing the message to the target user by adopting the message pushing mode.
Optionally, the pushing module 403 is further configured to: after the message is pushed to the target user, constructing a feature vector corresponding to each user according to the feature data of each user in the user group, the feature data of the video browsed by each user and the time sequence feature of the video browsed by each user;
and respectively calculating the similarity among the feature vectors, screening a preset number of communication users from the user group based on the sequence of the similarity from large to small, and establishing a communication association relation among the communication users.
According to the various embodiments, the technical means that the user group corresponding to the target video is determined according to the target video clicked by the target user, so that the video to be recommended is determined through the characteristic data of the user group and the video recommendation model, and the technical problems that the pushed video content is single and no distinction degree exists in the prior art are solved. The embodiment of the invention determines the video which is interested by the user by utilizing the characteristic data of the user group, is easier to attract the attention of the user, improves the click rate of the video, can enrich the content of the pushed video, avoids pushing a single video to the user, and therefore, the user requirement is better grasped.
It should be noted that, in the implementation of the apparatus for pushing video according to the present invention, the details of the method for pushing video are already described in detail above, and therefore, the repeated descriptions herein are not repeated.
Fig. 5 illustrates an exemplary system architecture 500 of a method or apparatus for pushing video to which embodiments of the present invention may be applied.
As shown in fig. 5, the system architecture 500 may include terminal devices 501, 502, 503, a network 504, and a server 505. The network 504 serves to provide a medium for communication links between the terminal devices 501, 502, 503 and the server 505. Network 504 may include various connection types, such as wired, wireless communication links, or fiber optic cables, to name a few.
A user may use terminal devices 501, 502, 503 to interact with a server 504 over a network 504 to receive or send messages, etc. The terminal devices 501, 502, 503 may have installed thereon various communication client applications, such as shopping-like applications, web browser applications, search-like applications, instant messaging tools, mailbox clients, social platform software, etc. (by way of example only).
The terminal devices 501, 502, 503 may be various electronic devices having a display screen and supporting web browsing, including but not limited to smart phones, tablet computers, laptop portable computers, desktop computers, and the like.
The server 505 may be a server providing various services, such as a background management server (for example only) providing support for shopping websites browsed by users using the terminal devices 501, 502, 503. The background management server may analyze and otherwise process the received data such as the item information query request, and feed back a processing result (for example, target push information, item information — just an example) to the terminal device.
It should be noted that the method for pushing a video provided by the embodiment of the present invention is generally performed by the server 505, and accordingly, the apparatus for pushing a video is generally disposed in the server 505. The method for pushing the video provided by the embodiment of the present invention may also be executed by the terminal devices 501, 502, and 503, and accordingly, the apparatus for pushing the video may be disposed in the terminal devices 501, 502, and 503.
It should be understood that the number of terminal devices, networks, and servers in fig. 5 is merely illustrative. There may be any number of terminal devices, networks, and servers, as desired for an implementation.
Referring now to FIG. 6, a block diagram of a computer system 600 suitable for use with a terminal device implementing embodiments of the present invention is shown. The terminal device shown in fig. 6 is only an example, and should not bring any limitation to the functions and the scope of use of the embodiments of the present invention.
As shown in fig. 6, the computer system 600 includes a Central Processing Unit (CPU) 601 that can perform various appropriate actions and processes according to a program stored in a Read Only Memory (ROM) 602 or a program loaded from a storage section 608 into a Random Access Memory (RAM) 603. In the RAM603, various programs and data necessary for the operation of the system 600 are also stored. The CPU 601, ROM 602, and RAM603 are connected to each other via a bus 604. An input/output (I/O) interface 605 is also connected to bus 604.
The following components are connected to the I/O interface 605: an input portion 606 including a keyboard, a mouse, and the like; an output portion 607 including a display such as a Cathode Ray Tube (CRT), a Liquid Crystal Display (LCD), and the like, and a speaker; a storage section 608 including a hard disk and the like; and a communication section 609 including a network interface card such as a LAN card, a modem, or the like. The communication section 609 performs communication processing via a network such as the internet. A driver 610 is also connected to the I/O interface 605 as needed. A removable medium 611 such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, or the like is mounted on the drive 610 as necessary, so that the computer program read out therefrom is mounted in the storage section 608 as necessary.
In particular, according to the embodiments of the present disclosure, the processes described above with reference to the flowcharts may be implemented as computer software programs. For example, embodiments of the present disclosure include a computer program article comprising a computer program embodied on a computer readable medium, the computer program comprising program code for performing the method illustrated by the flow chart. In such an embodiment, the computer program may be downloaded and installed from a network through the communication section 609 and/or installed from the removable medium 611. The computer program performs the above-described functions defined in the system of the present invention when executed by the Central Processing Unit (CPU) 601.
It should be noted that the computer readable medium shown in the present invention can be a computer readable signal medium or a computer readable storage medium or any combination of the two. A computer readable storage medium may be, for example, but not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination of the foregoing. More specific examples of the computer readable storage medium may include, but are not limited to: an electrical connection having one or more wires, a portable computer diskette, a hard disk, a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing. In the context of the present invention, a computer readable storage medium may be any tangible medium that can contain, or store a program for use by or in connection with an instruction execution system, apparatus, or device. In contrast, in the present invention, a computer readable signal medium may include a propagated data signal with computer readable program code embodied therein, for example, in baseband or as part of a carrier wave. Such a propagated data signal may take many forms, including, but not limited to, electro-magnetic, optical, or any suitable combination thereof. A computer readable signal medium may also be any computer readable medium that is not a computer readable storage medium and that can communicate, propagate, or transport a program for use by or in connection with an instruction execution system, apparatus, or device. Program code embodied on a computer readable medium may be transmitted using any appropriate medium, including but not limited to: wireless, wire, fiber optic cable, RF, etc., or any suitable combination of the foregoing.
The flowchart and block diagrams in the figures illustrate the architecture, functionality, and operation of possible implementations of systems, methods and computer program articles according to various embodiments of the present invention. In this regard, each block in the flowchart or block diagrams may represent a module, segment, or portion of code, which comprises one or more executable instructions for implementing the specified logical function(s). It should also be noted that, in some alternative implementations, the functions noted in the block may occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently, or the blocks may sometimes be executed in the reverse order, depending upon the functionality involved. It will also be noted that each block of the block diagrams or flowchart illustration, and combinations of blocks in the block diagrams or flowchart illustration, can be implemented by special purpose hardware-based systems that perform the specified functions or acts, or combinations of special purpose hardware and computer instructions.
The modules described in the embodiments of the present invention may be implemented by software or hardware. The described modules may also be provided in a processor, which may be described as: a processor includes a user group module, a determination module, and a push module, where the names of the modules do not in some cases constitute a limitation on the modules themselves.
As another aspect, the present invention also provides a computer-readable medium that may be contained in the apparatus described in the above embodiments; or may be separate and not incorporated into the device. The computer readable medium carries one or more programs which, when executed by a device, cause the device to comprise: determining a user group corresponding to a target video according to the target video clicked by a target user; determining a video to be recommended according to the characteristic data of the user group and a video recommendation model; pushing a message to the target user; the message carries an identifier of the video to be recommended; the video recommendation model is obtained by training a neural network based on training characteristic data of a video and corresponding label data.
According to the technical scheme of the embodiment of the invention, the user group corresponding to the target video is determined according to the target video clicked by the target user, so that the technical means of determining the video to be recommended is determined through the characteristic data of the user group and the video recommendation model, and the technical problems that the pushed video content is single and no distinction exists in the prior art are solved. The embodiment of the invention determines the video which is interested by the user by utilizing the characteristic data of the user group, is easier to attract the attention of the user, improves the click rate of the video, can enrich the content of the pushed video, avoids pushing a single video to the user, and therefore, the user requirement is better grasped.
The above-described embodiments should not be construed as limiting the scope of the invention. Those skilled in the art will appreciate that various modifications, combinations, sub-combinations, and substitutions can occur, depending on design requirements and other factors. Any modification, equivalent replacement, and improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.
Claims (10)
1. A method for pushing video, comprising:
determining a user group corresponding to a target video according to the target video clicked by a target user;
determining a video to be recommended according to the characteristic data of the user group and a video recommendation model;
pushing a message to the target user;
the message carries an identifier of the video to be recommended; the video recommendation model is obtained by training a neural network based on training characteristic data of a video and corresponding label data;
the training feature data of the video comprises: the characteristic data of the video, the characteristic data of the article corresponding to the video, and the characteristic data of the user who performs preset operation on the article.
2. The method of claim 1, wherein determining a video to be recommended according to the feature data of the user group and a video recommendation model comprises:
determining the characteristic data of the user group according to the user characteristic data of the user group;
inputting the characteristic data of the user group into a video recommendation model so as to output at least one video identifier and a corresponding preset operation probability;
and according to the preset operation probability, performing descending order on the at least one video, and screening a preset number of videos from the videos as videos to be recommended.
3. The method of claim 1, wherein determining a user group corresponding to a target video clicked by a target user comprises:
according to a target video clicked by a target user, taking each user concerning the target video as a user group corresponding to the target video; or,
determining a publishing user publishing the target video according to the target video clicked by the target user, and taking each user concerning the publishing user as a user group corresponding to the target video.
4. The method of claim 1, wherein the tag data of the video comprises: whether to execute preset operation on the article corresponding to the video and/or whether to execute preset operation on the video.
5. The method of claim 1, wherein pushing a message to the target user comprises:
determining a message pushing mode of the target user according to a message pushing model; the message pushing model is obtained by training a neural network based on message characteristic data and corresponding label data;
and pushing the message to the target user by adopting the message pushing mode.
6. The method of claim 1, further comprising, after pushing the message to the target user:
constructing a feature vector corresponding to each user according to the feature data of each user in the user group, the feature data of the video browsed by each user and the time sequence feature of the video browsed by each user;
and respectively calculating the similarity among the feature vectors, screening a preset number of communication users from the user group based on the sequence of the similarity from large to small, and establishing a communication association relationship among the communication users.
7. An apparatus for pushing video, comprising:
the user group module is used for determining a user group corresponding to a target video according to the target video clicked by a target user;
the determining module is used for determining a video to be recommended according to the characteristic data of the user group and the video recommendation model;
the pushing module is used for pushing a message to the target user;
the message carries an identifier of the video to be recommended; the video recommendation model is obtained by training a neural network based on training characteristic data of a video and corresponding label data;
the training feature data of the video comprises: the characteristic data of the video, the characteristic data of the article corresponding to the video, and the characteristic data of the user who performs preset operation on the article.
8. The apparatus of claim 7, wherein the determining module is further configured to:
determining the characteristic data of the user group according to the user characteristic data of the user group;
inputting the characteristic data of the user group into the video recommendation model to output at least one video identifier and a corresponding preset operation probability;
and according to the preset operation probability, performing descending order arrangement on the at least one video, and screening out a preset number of videos as videos to be recommended.
9. An electronic device, comprising:
one or more processors;
a storage device for storing one or more programs,
when executed by the one or more processors, cause the one or more processors to implement the method of any one of claims 1-6.
10. A computer-readable medium, on which a computer program is stored, which, when being executed by a processor, carries out the method according to any one of claims 1-6.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910822637.1A CN112449217B (en) | 2019-09-02 | 2019-09-02 | Method and device for pushing video, electronic equipment and computer readable medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910822637.1A CN112449217B (en) | 2019-09-02 | 2019-09-02 | Method and device for pushing video, electronic equipment and computer readable medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112449217A CN112449217A (en) | 2021-03-05 |
| CN112449217B true CN112449217B (en) | 2022-12-27 |
Family
ID=74734668
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910822637.1A Active CN112449217B (en) | 2019-09-02 | 2019-09-02 | Method and device for pushing video, electronic equipment and computer readable medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112449217B (en) |
Families Citing this family (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113190764B (en) * | 2021-03-09 | 2025-03-07 | 北京金山云网络技术有限公司 | Music recommendation method, device, equipment and computer-readable storage medium |
| CN115225702B (en) * | 2021-04-19 | 2023-08-08 | 北京达佳互联信息技术有限公司 | Information pushing method and device, electronic equipment and storage medium |
| CN114707041B (en) * | 2022-04-11 | 2023-12-01 | 中国电信股份有限公司 | Message recommendation method and device, computer readable medium and electronic equipment |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2011071819A (en) * | 2009-09-28 | 2011-04-07 | Kddi Corp | Recommendation server, recommendation determination method and program |
| CN103686236A (en) * | 2013-11-19 | 2014-03-26 | 乐视致新电子科技（天津）有限公司 | Method and system for recommending video resources |
| CN103731738A (en) * | 2014-01-23 | 2014-04-16 | 哈尔滨理工大学 | Video recommendation method and device based on user group behavioral analysis |
| CN107330115A (en) * | 2017-07-12 | 2017-11-07 | 广东工业大学 | A kind of information recommendation method and device |
| CN110012356A (en) * | 2019-04-16 | 2019-07-12 | 腾讯科技（深圳）有限公司 | Video recommendation method, device and equipment and computer storage medium |
| CN112905839A (en) * | 2021-02-10 | 2021-06-04 | 北京有竹居网络技术有限公司 | Model training method, model using device, storage medium and equipment |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US8037011B2 (en) * | 2008-09-15 | 2011-10-11 | Motorola Mobility, Inc. | Method and apparatus for recommending content items |
-
2019
- 2019-09-02 CN CN201910822637.1A patent/CN112449217B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2011071819A (en) * | 2009-09-28 | 2011-04-07 | Kddi Corp | Recommendation server, recommendation determination method and program |
| CN103686236A (en) * | 2013-11-19 | 2014-03-26 | 乐视致新电子科技（天津）有限公司 | Method and system for recommending video resources |
| CN103731738A (en) * | 2014-01-23 | 2014-04-16 | 哈尔滨理工大学 | Video recommendation method and device based on user group behavioral analysis |
| CN107330115A (en) * | 2017-07-12 | 2017-11-07 | 广东工业大学 | A kind of information recommendation method and device |
| CN110012356A (en) * | 2019-04-16 | 2019-07-12 | 腾讯科技（深圳）有限公司 | Video recommendation method, device and equipment and computer storage medium |
| CN112905839A (en) * | 2021-02-10 | 2021-06-04 | 北京有竹居网络技术有限公司 | Model training method, model using device, storage medium and equipment |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112449217A (en) | 2021-03-05 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109460513B (en) | Method and apparatus for generating click rate prediction model | |
| CN107577807B (en) | Method and device for pushing information | |
| CN109902849A (en) | User behavior prediction method and device, behavior prediction model training method and device | |
| CN113450172B (en) | Commodity recommendation method and device | |
| CN110119445B (en) | Method and device for generating feature vector and classifying text based on feature vector | |
| CN110020162B (en) | User identification method and device | |
| CN112449217B (en) | Method and device for pushing video, electronic equipment and computer readable medium | |
| CN110866040B (en) | User portrait generation method, device and system | |
| CN108595448B (en) | Information pushing method and device | |
| CN113742564B (en) | Method and device for pushing target resources | |
| CN112749323B (en) | Method and device for constructing user portrait | |
| CN113822734B (en) | Method and device for generating information | |
| CN109299351B (en) | Content recommendation method and device, electronic equipment and computer readable medium | |
| CN113792952B (en) | Method and apparatus for generating a model | |
| CN113269600B (en) | Information sending method and device | |
| CN113722624B (en) | Business icon display method and device | |
| CN113327145B (en) | Article recommendation method and device | |
| CN113327147A (en) | Method and device for displaying article information | |
| CN109408647B (en) | Method and apparatus for processing information | |
| CN112884538A (en) | Item recommendation method and device | |
| CN110163701B (en) | Method and device for pushing information | |
| CN113313542B (en) | Method and device for pushing channel pages | |
| CN111125502A (en) | Method and apparatus for generating information | |
| CN110968769A (en) | Method and equipment for pushing information and displaying information | |
| CN111723274B (en) | Method and device for processing information |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |