CN110971973A - Video pushing method and device and electronic equipment - Google Patents
Video pushing method and device and electronic equipment Download PDFInfo
- Publication number
- CN110971973A CN110971973A CN201911222506.6A CN201911222506A CN110971973A CN 110971973 A CN110971973 A CN 110971973A CN 201911222506 A CN201911222506 A CN 201911222506A CN 110971973 A CN110971973 A CN 110971973A
- Authority
- CN
- China
- Prior art keywords
- video
- user
- pushing
- vector
- target
- 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.)
- Pending
Links
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]
- H04N21/40—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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/70—Information retrieval; Database structures therefor; File system structures therefor of video data
- G06F16/73—Querying
- G06F16/735—Filtering based on additional data, e.g. user or group profiles
-
- 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]
- H04N21/40—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
- H04N21/4665—Learning process for intelligent management, e.g. learning user preferences for recommending movies characterized by learning algorithms involving classification methods, e.g. Decision trees
-
- 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)
- Multimedia (AREA)
- Signal Processing (AREA)
- General Health & Medical Sciences (AREA)
- Theoretical Computer Science (AREA)
- Social Psychology (AREA)
- Health & Medical Sciences (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Computational Linguistics (AREA)
- Physics & Mathematics (AREA)
- Computing Systems (AREA)
- Data Mining & Analysis (AREA)
- Computer Networks & Wireless Communication (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
- Two-Way Televisions, Distribution Of Moving Picture Or The Like (AREA)
Abstract
The embodiment of the invention provides a video pushing method and device and electronic equipment, and belongs to the field of computers. The method can acquire real-time behavior information generated when a target user watches videos; acquiring a target video matched with the real-time behavior information from a preset database, and pushing the target video for the target user; the preset database comprises the corresponding relation between the historical behavior information of the user and the videos which the user tends to watch. Therefore, the scheme of the invention can carry out video pushing aiming at the individuation of different users, improves the precision of video pushing, and solves the problem that the existing mode for pushing videos aiming at different group users has lower precision, thereby being incapable of meeting the individuation requirements of the users.
Description
Technical Field
The invention relates to the technical field of computers, in particular to a video pushing method and device and electronic equipment.
Background
When a user enjoys a video program through the internet, due to the fact that the types of videos are numerous, the video can be pushed to the user in order to save time and energy of the user for searching for the video program and improve the audience rating of the video. Among them, the current internet providers, especially video websites, often pursue centralized services for user groups, so as to push different types of videos for users of different groups.
However, each person as a single individual has special video watching requirements, and the existing method for pushing videos for users of different groups has low accuracy, so that the personalized requirements of the users cannot be met.
Disclosure of Invention
The invention provides a video pushing method, a video pushing device and electronic equipment, and aims to solve the problem that the existing mode for pushing videos for different groups of users is low in accuracy and cannot meet the personalized requirements of the users to a certain extent.
In a first aspect of the embodiments of the present invention, a video push method is provided, including:
acquiring real-time behavior information generated when a target user watches a video;
acquiring a target video matched with the real-time behavior information from a preset database, and pushing the target video for the target user;
the preset database comprises the corresponding relation between the historical behavior information of the user and the videos which the user tends to watch.
Optionally, the preset database includes user groups established in advance according to historical behavior information of users, characterization vectors of the user groups determined in advance, and characterization vectors of video IDs of videos that users tend to watch, where one user group corresponds to at least one video ID, and one video ID corresponds to at least one user group, the characterization vectors of the user groups indicate correlation between the user groups and the video IDs corresponding to the user groups, and the characterization vectors of the video IDs of videos that users tend to watch indicate correlation between the video IDs and the user groups corresponding to the video IDs;
the acquiring a target video matched with the real-time behavior information from a preset database and pushing the target video for the target user includes:
determining a target user group matched with the real-time behavior information of the target user according to a user group established in advance according to historical behavior information of the user;
according to the predetermined characterization vector of the user group, acquiring the characterization vector of the target user group;
and pushing the video for the target user according to the characterization vector of the target user group and the predetermined characterization vector of the video ID of the video which the user tends to watch.
Optionally, the pushing a video for the target user according to the characterization vector of the target user group and the predetermined characterization vector of the video ID of the video that the user tends to watch includes:
respectively calculating the vector similarity of the characterization vector of the target user group and the characterization vector of the video ID of the video which is determined by the user to tend to watch;
and pushing a video for the target user according to the vector similarity.
Optionally, the pushing a video for the target user according to the vector similarity includes:
sequencing the vector similarity, and pushing videos represented by the video IDs corresponding to the vector similarities in the preset number before the sequencing to the target user;
or
And pushing the video represented by the video ID corresponding to the vector similarity exceeding the preset value to the target user.
Optionally, the user group, the characterization vector of the user group, and the characterization vector of the video ID of the video that the user tends to watch are obtained by the following processes:
acquiring historical behavior information of a plurality of users and video IDs of videos which the users tend to watch;
constructing a sample sentence by using the historical behavior information of each user and the video ID of the video which tends to be watched, wherein one sample sentence comprises at least one user group and the video ID corresponding to the user group, and one user group comprises at least one word of the historical behavior information;
and training the sample sentences by adopting a word vector word2vec algorithm to obtain the characterization vectors of the user groups and the characterization vectors of the video IDs of the videos which the user tends to watch.
Optionally, the behavior information includes attribute information of the user and characteristic information of videos that tend to be watched, and the attribute information includes at least one of age, gender, and geographic location of the user.
In a second aspect of the embodiments of the present invention, there is provided a video push apparatus, including:
the information acquisition module is used for acquiring real-time behavior information generated when a target user watches videos;
the pushing module is used for acquiring a target video matched with the real-time behavior information from a preset database and pushing the target video for the target user;
the preset database comprises the corresponding relation between the historical behavior information of the user and the videos which the user tends to watch.
Optionally, the preset database includes user groups established in advance according to historical behavior information of users, characterization vectors of the user groups determined in advance, and characterization vectors of video IDs of videos that users tend to watch, where one user group corresponds to at least one video ID, and one video ID corresponds to at least one user group, the characterization vectors of the user groups indicate correlation between the user groups and the video IDs corresponding to the user groups, and the characterization vectors of the video IDs of videos that users tend to watch indicate correlation between the video IDs and the user groups corresponding to the video IDs;
the push module comprises:
the first determining submodule is used for determining a target user group matched with the real-time behavior information of the target user according to a user group established in advance according to historical behavior information of the user;
the first obtaining submodule is used for obtaining the representation vector of the target user group according to the representation vector of the predetermined user group;
and the pushing submodule is used for pushing the video for the target user according to the characterization vector of the target user group and the predetermined characterization vector of the video ID of the video which the user tends to watch.
Optionally, the pushing sub-module includes:
the calculation unit is used for calculating the vector similarity of the characterization vector of the target user group and the characterization vector of the video ID of the video which is determined by the user to tend to watch;
and the pushing unit is used for pushing a video for the target user according to the vector similarity.
Optionally, the pushing unit is specifically configured to:
sequencing the vector similarity, and pushing videos represented by the video IDs corresponding to the vector similarities in the preset number before the sequencing to the target user;
or
And pushing the video represented by the video ID corresponding to the vector similarity exceeding the preset value to the target user.
Optionally, the user group, the characterization vector of the user group, and the characterization vector of the video ID of the video that the user tends to watch are obtained by the following processes:
acquiring historical behavior information of a plurality of users and video IDs of videos which the users tend to watch;
constructing a sample sentence by using the historical behavior information of each user and the video ID of the video which tends to be watched, wherein one sample sentence comprises at least one user group and the video ID corresponding to the user group, and one user group comprises at least one word of the historical behavior information;
and training the sample sentences by adopting a word vector word2vec algorithm to obtain the characterization vectors of the user groups and the characterization vectors of the video IDs of the videos which the user tends to watch.
Optionally, the behavior information includes attribute information of the user and characteristic information of videos that tend to be watched, and the attribute information includes at least one of age, gender, and geographic location of the user.
In a third aspect of the embodiments of the present invention, there is further provided an electronic device, including a processor, a communication interface, a memory, and a communication bus, where the processor, the communication interface, and the memory complete communication with each other through the communication bus;
a memory for storing a computer program;
and the processor is used for realizing the steps of the video pushing method when executing the program stored in the memory.
In a fourth aspect of the embodiments of the present invention, there is also provided a computer-readable storage medium on which a computer program is stored, which when executed by a processor implements the video push method described above.
In a fifth aspect of the present invention, there is also provided a computer program product containing instructions which, when run on a computer, cause the computer to perform the video push method described above.
Aiming at the prior art, the invention has the following advantages:
according to the video pushing method provided by the embodiment of the invention, the database comprising the corresponding relation between the historical behavior information of the user and the videos which are watched in a tendency is established in advance, and the target video matched with the real-time behavior information of the target user watching the videos is obtained in the database and pushed to the target user.
The foregoing description is only an overview of the technical solutions of the present invention, and the embodiments of the present invention are described below in order to make the technical means of the present invention more clearly understood and to make the above and other objects, features, and advantages of the present invention more clearly understandable.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below.
Fig. 1 is a flowchart illustrating steps of a video pushing method according to an embodiment of the present invention;
fig. 2 is a flowchart illustrating steps of another video pushing method according to an embodiment of the present invention;
FIG. 3 is a schematic diagram illustrating a principle of training a sample sentence by using a word2vec algorithm according to an embodiment of the present invention;
fig. 4 is a block diagram of a video pushing apparatus according to an embodiment of the present invention;
fig. 5 is a block diagram of another video pushing apparatus according to an embodiment of the present invention;
fig. 6 is a block diagram of an electronic device provided in an embodiment of the present invention.
Detailed Description
Exemplary embodiments of the present invention will be described in more detail below with reference to the accompanying drawings.
Fig. 1 is a flowchart of steps of a video pushing method according to an embodiment of the present invention, as shown in fig. 1, the method may include:
Optionally, the real-time behavior information may further include operation information of the video watched by the user in the process of watching the video, and for example, the operation information may be one or a combination of the following: like operations, collection operations, forwarding operations, fast forward viewing operations, etc.
For example, if a user clicks the video B to view the video B and forwards the video B during the viewing process, the real-time behavior information of the user viewing the video B is recorded in real time during the viewing process of the user viewing the video B, for example, the feature information of the video B and the forwarding operation information of the user viewing the video B during the viewing process of the video B are recorded.
In the embodiment of the invention, the target user is any user, so that in the embodiment of the invention, when any user watches the video, the real-time behavior information generated when the user watches the video can be acquired, and the video is pushed for the user.
Therefore, in the embodiment of the invention, when the video is pushed to the user, the actual watching behavior of the video by the user can be referred to, so that the pushed video is matched with the actual watching behavior of the video by the user, the pushing accuracy of the video is improved, and the click rate of the video is further improved.
In addition, optionally, the embodiment may be specifically applied to a scene in which the user watches the short video, and since the user can generate more behavior information in a unit time in the process of watching the short video, the behavior information generated by the user can be better pushed to the user. Therefore, the duration of the video watched by the user may be less than a preset value, for example, the duration may be a short video with a duration of any duration within 10 seconds to two minutes.
And 102, acquiring a target video matched with the real-time behavior information from a preset database, and pushing the target video for the target user.
The preset database comprises the corresponding relation between the historical behavior information of the user and the videos which the user tends to watch. The historical behavior information may include characteristic information of videos that the user tends to watch. Optionally, the historical behavior information may further include operation information of the video watched by the user during the process of watching the video.
The target video is pushed for the target user, namely, the target video is displayed to the target user, so that the target user can view the video meeting the actual viewing requirement of the target user without searching by the target user.
In addition, the pushed video can be displayed after the user refreshes the page, or can be displayed after the video currently watched by the user is pushed, that is, the playing interface of the video currently watched by the user is closed.
For example, the user a is watching the short video B in a certain short video APP and has forwarded the short video B, and then the real-time behavior information (including the feature information of the short video B and the forwarding operation information of the user a on the short video B) of the video watched by the user a can be obtained at this time, so that the video matched with the above-mentioned real-time behavior information of the user a can be selected in the preset database for pushing, that is, the video matched with both the feature information and the forwarding operation information of the short video B is selected for pushing, and therefore, after the user a refreshes the page, the video to be pushed can be automatically displayed, so that the user a can watch the video meeting the actual watching requirement of the user a without searching by the user a.
In summary, in the video pushing method provided in the embodiment of the present invention, a database including a correspondence between the user historical behavior information and the videos that tend to be watched is pre-established, and the target video matched with the real-time behavior information of the video watched by the target user is obtained from the database and pushed to the target user. Therefore, according to the embodiment of the invention, the video is pushed by combining the corresponding relation between the historical behavior information of the user watching the video and the video which tends to be watched and the real-time behavior information of the target user waiting for pushing the video, so that the personalized video pushing can be realized aiming at the watching behaviors of different users when watching the video, the video pushing precision can be improved, and the problems that the precision of the existing video pushing mode aiming at different groups of users is low and the personalized requirements of the users cannot be met are solved.
Fig. 2 is a flowchart of steps of another video pushing method provided in an embodiment of the present invention, and as shown in fig. 2, the method may include:
Optionally, the real-time behavior information may include attribute information of the user in addition to related operations such as forwarding generated when the user watches the video and the like included in the above embodiment, and feature information of the video watched by the user.
The user attribute information is the inherent attribute of the user, and comprises at least one of age, gender and the geographic position of the user. It is to be understood that the specific content included in the user attribute information is not limited thereto. For example, when users of different ages and different genders are in different geographical positions, videos to be watched often have larger differences, and in the embodiment of the invention, when the videos are pushed to the users, the inherent attributes of the users can be referred to, so that the pushed videos can conform to the inherent attributes of the users, the video pushing accuracy is improved, and the click rate of the videos is further improved.
In addition, the above feature information is the type of the video content. For example, a video website generally creates a label for a provided video, for example, a label for a movie video is "movie", a label for a laugh video is "laugh", a label for a music video is "music", and a label for a life video is "life", so that feature information of the video can be represented by a video label.
For example, a certain user a clicks a video B to view the video B and forwards the video B during the viewing process, and during the viewing process of the video B, real-time behavior information of the user a viewing the video B is recorded in real time, for example, attribute information of the user a, feature information of the video B, and forwarding operation information of the user a viewing the video B during the viewing process of the video B are recorded.
In addition, in the embodiment of the present invention, the target user is any user, and therefore, in the embodiment of the present invention, when any user watches a video, the user attribute information of the user, the feature information of the video watched at that time, and the video operation information may be acquired, so as to perform video push for the user.
The preset database comprises user groups established in advance according to historical behavior information of users, characterization vectors of the user groups determined in advance and characterization vectors of video IDs of videos which users tend to watch, wherein one user group corresponds to at least one video ID, one video ID corresponds to at least one user group, the characterization vectors of the user groups represent the correlation between the user groups and the video IDs corresponding to the user groups, and the characterization vectors of the video IDs of the videos which users tend to watch represent the correlation between the video IDs and the user groups corresponding to the video IDs.
Similarly, the historical behavior information may include attribute information of the user in addition to the feature information of the video that the user intends to view and related user operations such as forwarding generated when the user views the video, which are included in the above embodiments.
For example, the user a is watching the short video B in a certain short video APP and has forwarded the short video B, and then the real-time behavior information (including the attribute information of the user a, the feature information of the short video B, and the forwarding operation information of the user a on the short video B) of the video watched by the user a can be obtained at this time, so that the video matched with the real-time behavior information of the user a can be selected in the preset database for pushing, that is, the video matched with all of the attribute information of the user a, the feature information of the short video B, and the forwarding operation information is selected for pushing, so that the video to be pushed can be automatically displayed after the user a refreshes the page, and the user a can watch the video meeting the actual watching requirement of the user a without searching by the user a.
In the embodiment of the invention, when the historical behavior information comprises the attribute information of the user and the characteristic information of the video which the user tends to watch, the user attribute information of a large number of users and the characteristic information of the video which the user tends to watch are collected in advance to establish the user group. For example, when the user attribute information includes age, gender, and geographic location of the user, if the collected video watching list of the user's tendency to watch in one day is: the method comprises the following steps of establishing a first video, a second video, a third video and a fourth video, wherein the video characteristic information of the first video is 'movie & TV', the video characteristic information of the second video is 'make up', the video characteristic information of the third video is 'music', and the video characteristic information of the fourth video is 'life', and aiming at the information of the user, the established user groups are as follows:
a first user group: age _2_ gender _1_ geographic location _0_ 0;
a second user group: age _2_ gender _1_ geographic location _ shanghai _ movie _ 0;
and (3) grouping the third users: age _2_ gender _1_ geographic location _ shanghai _ movie _ makes a fun;
fourth user grouping: age _2_ gender _1_ geographical location _ shanghai _ laugh _ music;
grouping the fifth users: age _2_ gender _1_ geographical location _ shanghai _ music _ life.
Wherein age _2, represents a specific age; sex _1, which represents a specific sex, male or female; and the geographic position _0 represents that the user does not watch the video, and the specific content of the geographic position is marked as '0'.
As can be seen from the above, when the user group is established, the feature information of the video that the user tends to watch in the preset time period can be distinguished according to the feature information of the video before and after the click behavior. For example, if the user watches the "movie" video before a certain click action and watches the "make a fun" video after the click action, the obtained user group is the third user group, i.e., age _2_ gender _1_ geographical location _ shanghai _ movie _ make a fun.
Therefore, in the embodiment of the invention, although the user is the same user, the user belongs to different user groups before/after different click behaviors. For example, the initial state is: age _2_ gender _1_ geographic location _0_0, after watching a movie video, will become: age _2_ gender _1_ geographic location _ shanghai _ movie _ 0.
In addition, when the user _ B clicks the fifth video, the Apache Kafka real-time stream sends a video click record, so that the user ID finds the user attribute information "age _2_ gender _1_ geographical location _ shanghai" of the user _ B, and obtains the feature information "movie" of the fifth video, so that the target user group matching the user attribute information of the target user and the feature of the video watched at the current time can be determined as the second user group, i.e., age _2_ gender _1_ geographical location _ shanghai _ movie _0, from a plurality of user groups obtained in advance. Apache Kafka is an open source message system project, written by Scala. The goal of this project is to provide a uniform, high throughput, low latency platform for processing real-time data.
And 203, acquiring the characterization vector of the target user group according to the predetermined characterization vector of the user group.
In the embodiment of the invention, one user group corresponds to at least one video ID, and the characterization vector of the user group represents the correlation between the user group and the video ID corresponding to the user group. And if the user group comprises at least one user attribute word and at least one video characteristic word, the video ID corresponding to the user group is the ID of the video described by the video characteristic word included in the user group.
And step 204, pushing the video for the target user according to the characterization vector of the target user group and the predetermined characterization vector of the video ID of the video which the user tends to watch.
In the embodiment of the invention, one video ID corresponds to at least one user group, and the characterization vector of the video ID represents the correlation of the user group corresponding to the video ID. And the user group comprises at least one user attribute word and at least one video characteristic word, and the user group corresponding to the video ID is the user group to which the video characteristic word of the video represented by the video ID belongs.
In step 204, after the characterization vector of the target user group is obtained, video pushing may be performed on the target user according to the obtained characterization vector of the target user group and the predetermined characterization vector of the video ID of the video that the user tends to watch.
Optionally, the user group, the characterization vector of the user group, and the characterization vector of the video ID of the video that the user tends to watch are obtained by the following processes:
acquiring historical behavior information of a plurality of users and video IDs of videos which the users tend to watch;
constructing a sample sentence by using the historical behavior information of each user and the video ID of the video which tends to be watched, wherein one sample sentence comprises at least one user group and the video ID corresponding to the user group, and one user group comprises at least one word of the historical behavior information;
and training the sample sentences by adopting a word2vec algorithm to obtain the characterization vectors of the user groups and the characterization vectors of the video IDs of the videos which the user tends to watch.
For an example of the process of establishing the user group, refer to the description after step 203, which is not described herein again. The process of obtaining the token vector for the user group and the token vector for the video ID is as follows:
when the user attribute information includes age, gender and geographic location of the user, if a video watching list collected about the tendency of the user to watch in one day is: the method comprises the following steps of converting the behaviors of a user into sample sentences as follows, wherein the video characteristic information of a first video is 'movie', the video characteristic information of a second video is 'make up', the video characteristic information of a third video is 'music', and the video characteristic information of a fourth video is 'life':
age _2_ gender _1_ geographic location _0_0, video _ ID1, age _2_ gender _1_ geographic location _ shanghai _ movie _0, video _ ID2, age _2_ gender _1_ geographic location _ shanghai _ movie _ laugh, video _ ID3, age _2_ gender _1_ geographic location _ shanghai _ laugh _ music, video _ ID4, fifth user packet: age _2_ gender _1_ geographical location _ shanghai _ music _ life.
Wherein ID1 is the ID of the first video, ID2 is the ID of the second video, ID3 is the ID of the third video, and ID4 is the ID of the fourth video.
As can be seen from the above, in the embodiment of the present invention, a sample sentence is generated from a video watching behavior that a user tends to watch in a preset time period, and in order to obtain a characterization vector of a user group and a characterization vector of a video ID that the user tends to watch, a large number of the sample sentences need to be obtained.
In addition, word2vec can be trained efficiently on millions of orders of magnitude dictionaries and billions of datasets; secondly, the training result, namely the word vector, obtained by the tool can well measure the similarity between words.
Further alternatively, a skip-gram model in word2vec may be used.
In the embodiment of the invention, when the word2vec algorithm is applied to the process of determining the characterization vector of the user group and the characterization vector of the video ID which tends to be watched, the obtained massive sample sentences only need to be trained by adopting the word2vec algorithm, and the characterization vector of the user group and the characterization vector of the video ID which tends to be watched can be obtained. Wherein, optionally, the token vector of the user group and the token vector of the video ID prone to be watched are 64-dimensional vectors respectively.
Specifically, a group of character string groups in which a user grouping character string (i.e., a user attribute word and a video feature word corresponding to one user grouping) and a video ID which tends to be viewed are connected in series is taken as a sentence in text processing; and (3) predicting other words in the sentence as output (namely other groups or video IDs) by taking any word in the word2vec as input (in this case, a group character string or a video ID), and training a fully-connected neural network. After a large number of samples of back propagation training, the fully connected neural network can produce 64-dimensional characterization vectors of all grouping strings and video IDs inclined to view.
In summary, as shown in fig. 3, in the embodiment of the present invention, a large amount of user attribute information of users, feature information of videos that users tend to watch in a preset time period, and a video ID are collected in advance and stored in a Couchbase (distributed cache system), so that according to the data stored in the Couchbase, a feature vector of a user group and a feature vector of a video ID that users tend to watch are obtained, and when a real-time video watching behavior of a certain user is obtained, a feature vector of a target user group that is matched with the user attribute information of the user and the real-time video watching behavior is determined, and then video is pushed to the target user according to the feature vector of the target user group and the feature vector of the video ID that users tend to watch.
Namely, in the embodiment of the invention, different user groups and characterization vectors of video IDs prone to be watched are trained offline in the same dimension, a real-time stream task is developed, and the corresponding characterization vector is found according to the user attribute information of each user and the dynamic watching behavior of the video, so that the video most similar to the user is found for recalling into the sequencing model.
Therefore, the embodiment of the invention combines the user attribute information of the user with the actual behavior of watching the video in real time, can carry out video pushing aiming at the individuation of different users, improves the accuracy of video pushing, and solves the problem that the existing mode for pushing the video aiming at different group users has lower accuracy, thereby being incapable of meeting the individuation requirements of the users.
In addition, in the embodiment of the invention, the video is pushed to the target user, namely, the matched video is displayed to the target user according to the user attribute information and the actual video watching behavior of the target user, so that the target user can watch the video which accords with the user attribute information and the actual watching behavior of the target user without searching by himself.
In addition, the pushed video can be displayed after the user refreshes the page, or can be displayed after the video currently watched by the user is pushed, that is, the playing interface of the currently watched video is closed.
Optionally, the pushing a video for the target user according to the characterization vector of the target user group and the predetermined characterization vector of the video ID of the video that the user tends to watch includes:
respectively calculating the vector similarity of the characterization vector of the target user group and the characterization vector of the video ID of the video which is determined by the user to tend to watch;
and pushing a video for the target user according to the vector similarity.
The vector similarity can be calculated by using a Pearson correlation coefficient method, a Brazilian distance method, a Manhattan distance method and the like.
Further, optionally, the pushing a video for the target user according to the vector similarity includes:
sequencing the vector similarity, and pushing videos represented by the video IDs corresponding to the vector similarities in the preset number before the sequencing to the target user;
or
And pushing the video represented by the video ID corresponding to the vector similarity exceeding the preset value to the target user.
The larger the vector similarity is, the more the video represented by the video ID corresponding to the vector similarity meets the user attribute information and the actual watching demand of the user. Therefore, the vector similarities are sorted, and videos represented by the video IDs corresponding to the preset number of vector similarities before the sorting are pushed to the target user, or videos represented by the video IDs corresponding to the vector similarities exceeding the preset value are pushed to the target user, so that the precision of the pushed videos can be further improved.
In addition, for the video push method of the embodiment of the present invention, after ab test, the test indexes are as follows:
a full-platform new user: PPUI +0.114, UCTR-0.028PP, retention +0.134PP, forward interaction rate +0.02PP, display mean age (days) -46.57, GINI-0.02827;
android new user: PPUI +0.117, UCTR-0.059PP, retention +0.122PP, forward interaction Rate +0.025PP, display average age (days) -51.63, GINI-0.03856;
IOS (apple mobile operating system) new user: PPUI +0.096, UCTR +0.087PP, Retention +0.185PP, Forward interactivity-0.021 PP, display average age (days) -12.67, GINI-0.0169.
Wherein, PPUI represents the user click duration, UCTR represents the user click rate, PP represents the percentage, and GINI represents the user diversity.
Therefore, the click time of the user is obviously prolonged, and the user retention is greatly improved, namely the user experience is improved. Meanwhile, the video age is reduced, the diversity is improved, and the results have positive effects on business ecology.
In summary, according to the video pushing method provided by the embodiment of the present invention, the user attribute information of the user and the feature information of the video watched at the current time are obtained, and based on the obtained user attribute information, the video is pushed for the user, and the user attribute information of the user and the actual behavior of watching the video in real time are combined, so that video pushing can be performed for individuation of different users, the video pushing accuracy is improved, and the problem that the existing video pushing manner for users of different groups has low accuracy is solved, so that the individuation requirement of the user cannot be met.
In addition, it should be noted that, in the embodiment of the present invention, taking the above behavior information including the attribute information of the user and the feature information of the video as an example, the process of establishing the preset database and selecting the video to be pushed from the preset database according to the real-time behavior information generated when the target user watches the video (i.e. the attribute information of the target user and the feature information of the video being watched by the target user) is elaborated. For the specific implementation when the behavior information includes other information (for example, operation information when the user watches a video), the above process may be referred to, and details are not described here.
Fig. 4 is a block diagram of a video push apparatus according to an embodiment of the present invention, and as shown in fig. 4, the apparatus 60 may include:
the information acquisition module 601 is configured to acquire real-time behavior information generated when a target user watches a video;
a pushing module 602, configured to obtain a target video matched with the real-time behavior information from a preset database, and push the target video for the target user;
the preset database comprises the corresponding relation between the historical behavior information of the user and the videos which the user tends to watch.
In summary, the video pushing apparatus provided in the embodiment of the present invention pre-establishes the database including the corresponding relationship between the historical behavior information of the user and the videos that are intended to be watched, and obtains the target video that matches the real-time behavior information of the target user watching the videos from the database and pushes the target video to the target user, so that, in the embodiment of the present invention, the video pushing is performed by combining the corresponding relationship between the historical behavior information of the user watching the videos and the videos that are intended to be pushed and the real-time behavior information of the target user waiting to push the videos, thereby realizing personalized video pushing for different users, improving the accuracy of video pushing, and solving the problem that the accuracy of the existing video pushing method for users of different groups is low, so that the personalized requirements of the users cannot be met.
Fig. 5 is a block diagram of another video push apparatus according to an embodiment of the present invention, and as shown in fig. 5, the video push apparatus 70 may include:
the information acquisition module 701 is used for acquiring real-time behavior information generated when a target user watches videos;
a pushing module 702, configured to obtain a target video matched with the real-time behavior information from a preset database, and push the target video for the target user;
the preset database comprises the corresponding relation between the historical behavior information of the user and the videos which the user tends to watch.
Optionally, the preset database includes user groups established in advance according to historical behavior information of users, characterization vectors of the user groups determined in advance, and characterization vectors of video IDs of videos that users tend to watch, where one user group corresponds to at least one video ID, and one video ID corresponds to at least one user group, the characterization vectors of the user groups indicate correlation between the user groups and the video IDs corresponding to the user groups, and the characterization vectors of the video IDs of videos that users tend to watch indicate correlation between the video IDs and the user groups corresponding to the video IDs;
as shown in fig. 5, the pushing module 702 includes:
a first determining sub-module 7021, configured to determine, according to a user group established in advance according to historical behavior information of a user, a target user group matched with the real-time behavior information of the target user;
a first obtaining sub-module 7022, configured to obtain a characterization vector of the target user group according to a predetermined characterization vector of the user group;
and the pushing submodule 7023 is configured to push a video to the target user according to the characterization vector of the target user group and a predetermined characterization vector of a video ID of a video that the user tends to watch.
Optionally, the pushing sub-module 7023 includes:
the calculation unit is used for calculating the vector similarity of the characterization vector of the target user group and the characterization vector of the video ID of the video which is determined by the user to tend to watch;
and the pushing unit is used for pushing a video for the target user according to the vector similarity.
Optionally, the pushing unit is specifically configured to:
sequencing the vector similarity, and pushing videos represented by the video IDs corresponding to the vector similarities in the preset number before the sequencing to the target user;
or
And pushing the video represented by the video ID corresponding to the vector similarity exceeding the preset value to the target user.
Optionally, the user group, the characterization vector of the user group, and the characterization vector of the video ID of the video that the user tends to watch are obtained by the following processes:
acquiring historical behavior information of a plurality of users and video IDs of videos which the users tend to watch;
constructing a sample sentence by using the historical behavior information of each user and the video ID of the video which tends to be watched, wherein one sample sentence comprises at least one user group and the video ID corresponding to the user group, and one user group comprises at least one word of the historical behavior information;
and training the sample sentences by adopting a word vector word2vec algorithm to obtain the characterization vectors of the user groups and the characterization vectors of the video IDs of the videos which the user tends to watch.
Optionally, the behavior information includes attribute information of the user and characteristic information of videos that tend to be watched, and the attribute information includes at least one of age, gender, and geographic location of the user.
In summary, the video pushing device provided in the embodiment of the present invention, by acquiring the user attribute information of the user and the target video watched at the current time, and pushing the video for the user according to the user attribute information and the target video, and combining the user attribute information of the user and the actual behavior of watching the video in real time, can perform video pushing for personalization of different users, improve the accuracy of video pushing, and solve the problem that the accuracy of the existing video pushing manner for users of different groups is low, so that the personalized requirements of the users cannot be met.
An embodiment of the present invention further provides an electronic device, as shown in fig. 6, including a processor 801, a communication interface 802, a memory 803, and a communication bus 804, where the processor 801, the communication interface 802, and the memory 803 complete mutual communication through the communication bus 804;
a memory 803 for storing a computer program;
the processor 801 is configured to implement the following steps when executing the program stored in the memory 803:
acquiring real-time behavior information generated when a target user watches a video;
acquiring a target video matched with the real-time behavior information from a preset database, and pushing the target video for the target user;
the preset database comprises the corresponding relation between the historical behavior information of the user and the videos which the user tends to watch.
Optionally, the preset database includes user groups established in advance according to historical behavior information of users, characterization vectors of the user groups determined in advance, and characterization vectors of video IDs of videos that users tend to watch, where one user group corresponds to at least one video ID, and one video ID corresponds to at least one user group, the characterization vectors of the user groups indicate correlation between the user groups and the video IDs corresponding to the user groups, and the characterization vectors of the video IDs of videos that users tend to watch indicate correlation between the video IDs and the user groups corresponding to the video IDs;
the acquiring a target video matched with the real-time behavior information from a preset database and pushing the target video for the target user includes:
determining a target user group matched with the real-time behavior information of the target user according to a user group established in advance according to historical behavior information of the user;
according to the predetermined characterization vector of the user group, acquiring the characterization vector of the target user group;
and pushing the video for the target user according to the characterization vector of the target user group and the predetermined characterization vector of the video ID of the video which the user tends to watch.
Optionally, the pushing a video for the target user according to the characterization vector of the target user group and the predetermined characterization vector of the video ID of the video that the user tends to watch includes:
respectively calculating the vector similarity of the characterization vector of the target user group and the characterization vector of the video ID of the video which is determined by the user to tend to watch;
and pushing a video for the target user according to the vector similarity.
Optionally, the pushing a video for the target user according to the vector similarity includes:
sequencing the vector similarity, and pushing videos represented by the video IDs corresponding to the vector similarities in the preset number before the sequencing to the target user;
or
And pushing the video represented by the video ID corresponding to the vector similarity exceeding the preset value to the target user.
Optionally, the user group, the characterization vector of the user group, and the characterization vector of the video ID of the video that the user tends to watch are obtained by the following processes:
acquiring historical behavior information of a plurality of users and video IDs of videos which the users tend to watch;
constructing a sample sentence by using the historical behavior information of each user and the video ID of the video which tends to be watched, wherein one sample sentence comprises at least one user group and the video ID corresponding to the user group, and one user group comprises at least one word of the historical behavior information;
and training the sample sentences by adopting a word vector word2vec algorithm to obtain the characterization vectors of the user groups and the characterization vectors of the video IDs of the videos which the user tends to watch.
Optionally, the behavior information includes attribute information of the user and characteristic information of videos that tend to be watched, and the attribute information includes at least one of age, gender, and geographic location of the user.
The communication bus mentioned in the above terminal may be a Peripheral Component Interconnect (PCI) bus, an Extended Industry Standard Architecture (EISA) bus, or the like. The communication bus may be divided into an address bus, a data bus, a control bus, etc. For ease of illustration, only one thick line is shown, but this does not mean that there is only one bus or one type of bus.
The communication interface is used for communication between the terminal and other equipment.
The Memory may include a Random Access Memory (RAM) or a non-volatile Memory (non-volatile Memory), such as at least one disk Memory. Optionally, the memory may also be at least one memory device located remotely from the processor.
The Processor may be a general-purpose Processor, and includes a Central Processing Unit (CPU), a Network Processor (NP), and the like; the device can also be a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), a Field Programmable Gate Array (FPGA) or other Programmable logic device, a discrete Gate or transistor logic device, or a discrete hardware component.
In yet another embodiment of the present invention, a computer-readable storage medium is further provided, which stores instructions that, when executed on a computer, cause the computer to execute the video pushing method described in any of the above embodiments.
In yet another embodiment provided by the present invention, there is also provided a computer program product containing instructions which, when run on a computer, cause the computer to perform the video push method described in the above embodiment.
In the above embodiments, the implementation may be wholly or partially realized by software, hardware, firmware, or any combination thereof. When implemented in software, may be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. When loaded and executed on a computer, cause the processes or functions described in accordance with the embodiments of the invention to occur, in whole or in part. The computer may be a general purpose computer, a special purpose computer, a network of computers, or other programmable device. The computer instructions may be stored in a computer readable storage medium or transmitted from one computer readable storage medium to another, for example, from one website site, computer, server, or data center to another website site, computer, server, or data center via wired (e.g., coaxial cable, fiber optic, Digital Subscriber Line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.). The computer-readable storage medium can be any available medium that can be accessed by a computer or a data storage device, such as a server, a data center, etc., that incorporates one or more of the available media. The usable medium may be a magnetic medium (e.g., floppy Disk, hard Disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium (e.g., Solid State Disk (SSD)), among others.
It is noted that, herein, relational terms such as first and second, and the like may be used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Also, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other identical elements in a process, method, article, or apparatus that comprises the element.
All the embodiments in the present specification are described in a related manner, and the same and similar parts among the embodiments may be referred to each other, and each embodiment focuses on the differences from the other embodiments. In particular, for the system embodiment, since it is substantially similar to the method embodiment, the description is simple, and for the relevant points, reference may be made to the partial description of the method embodiment.
The above description is only for the preferred embodiment of the present invention, and is not intended to limit the scope of the present invention. Any modification, equivalent replacement, or improvement made within the spirit and principle of the present invention shall fall within the protection scope of the present invention.
Claims (14)
1. A video push method, comprising:
acquiring real-time behavior information generated when a target user watches a video;
acquiring a target video matched with the real-time behavior information from a preset database, and pushing the target video for the target user;
the preset database comprises the corresponding relation between the historical behavior information of the user and the videos which the user tends to watch.
2. The video pushing method according to claim 1, wherein the preset database includes user groups established in advance according to historical behavior information of users, predetermined characterization vectors of the user groups, and predetermined characterization vectors of video IDs of videos that users tend to watch, where one user group corresponds to at least one video ID and one video ID corresponds to at least one user group, the characterization vectors of the user groups represent the correlation between the user groups and the video IDs corresponding to the user groups, and the characterization vectors of the video IDs of videos that users tend to watch represent the correlation between the video IDs and the user groups corresponding to the video IDs;
the acquiring a target video matched with the real-time behavior information from a preset database and pushing the target video for the target user includes:
determining a target user group matched with the real-time behavior information of the target user according to a user group established in advance according to historical behavior information of the user;
according to the predetermined characterization vector of the user group, acquiring the characterization vector of the target user group;
and pushing the video for the target user according to the characterization vector of the target user group and the predetermined characterization vector of the video ID of the video which the user tends to watch.
3. The video pushing method according to claim 2, wherein the pushing of the video for the target user according to the characterization vector of the target user group and the characterization vector of the video ID of the video that the predetermined user tends to watch comprises:
respectively calculating the vector similarity of the characterization vector of the target user group and the characterization vector of the video ID of the video which is determined by the user to tend to watch;
and pushing a video for the target user according to the vector similarity.
4. The video pushing method according to claim 3, wherein said pushing a video for the target user according to the vector similarity comprises:
sequencing the vector similarity, and pushing videos represented by the video IDs corresponding to the vector similarities in the preset number before the sequencing to the target user;
or
And pushing the video represented by the video ID corresponding to the vector similarity exceeding the preset value to the target user.
5. The video pushing method according to claim 2, wherein the user group, the characterization vector of the user group, and the characterization vector of the video ID of the video that the user tends to watch are obtained by:
acquiring historical behavior information of a plurality of users and video IDs of videos which the users tend to watch;
constructing a sample sentence by using the historical behavior information of each user and the video ID of the video which tends to be watched, wherein one sample sentence comprises at least one user group and the video ID corresponding to the user group, and one user group comprises at least one word of the historical behavior information;
and training the sample sentences by adopting a word vector word2vec algorithm to obtain the characterization vectors of the user groups and the characterization vectors of the video IDs of the videos which the user tends to watch.
6. The video pushing method according to any one of claims 1 to 5, wherein the behavior information includes attribute information of the user and characteristic information of videos that tend to be watched, and the attribute information includes at least one of age, gender, and geographical location of the user.
7. A video push apparatus, comprising:
the information acquisition module is used for acquiring real-time behavior information generated when a target user watches videos;
the pushing module is used for acquiring a target video matched with the real-time behavior information from a preset database and pushing the target video for the target user;
the preset database comprises the corresponding relation between the historical behavior information of the user and the videos which the user tends to watch.
8. The video pushing apparatus according to claim 7, wherein the preset database includes user groups pre-established according to historical behavior information of users, pre-determined characterization vectors of the user groups, and pre-determined characterization vectors of video IDs of videos that users tend to watch, where one user group corresponds to at least one video ID and one video ID corresponds to at least one user group, the characterization vectors of the user groups represent correlations between the user groups and the video IDs corresponding to the user groups, and the characterization vectors of the video IDs of videos that users tend to watch represent correlations between the video IDs and the user groups corresponding to the video IDs;
the push module comprises:
the first determining submodule is used for determining a target user group matched with the real-time behavior information of the target user according to a user group established in advance according to historical behavior information of the user;
the first obtaining submodule is used for obtaining the representation vector of the target user group according to the representation vector of the predetermined user group;
and the pushing submodule is used for pushing the video for the target user according to the characterization vector of the target user group and the predetermined characterization vector of the video ID of the video which the user tends to watch.
9. The video push apparatus according to claim 8, wherein the push sub-module comprises:
the calculation unit is used for calculating the vector similarity of the characterization vector of the target user group and the characterization vector of the video ID of the video which is determined by the user to tend to watch;
and the pushing unit is used for pushing a video for the target user according to the vector similarity.
10. The video pushing device according to claim 9, wherein the pushing unit is specifically configured to:
sequencing the vector similarity, and pushing videos represented by the video IDs corresponding to the vector similarities in the preset number before the sequencing to the target user;
or
And pushing the video represented by the video ID corresponding to the vector similarity exceeding the preset value to the target user.
11. The video pushing apparatus according to claim 8, wherein the user group, the characterization vector of the user group, and the characterization vector of the video ID of the video that the user tends to watch are obtained by:
acquiring historical behavior information of a plurality of users and video IDs of videos which the users tend to watch;
constructing a sample sentence by using the historical behavior information of each user and the video ID of the video which tends to be watched, wherein one sample sentence comprises at least one user group and the video ID corresponding to the user group, and one user group comprises at least one word of the historical behavior information;
and training the sample sentences by adopting a word vector word2vec algorithm to obtain the characterization vectors of the user groups and the characterization vectors of the video IDs of the videos which the user tends to watch.
12. The video pushing apparatus according to any one of claims 7 to 11, wherein the behavior information includes attribute information of the user and characteristic information of videos that tend to be watched, and the attribute information includes at least one of age, gender, and geographical location of the user.
13. An electronic device is characterized by comprising a processor, a communication interface, a memory and a communication bus, wherein the processor and the communication interface are used for realizing mutual communication by the memory through the communication bus;
a memory for storing a computer program;
a processor for implementing the steps of the video push method of any of claims 1-6 when executing a program stored in the memory.
14. A computer-readable storage medium, on which a computer program is stored, which, when being executed by a processor, carries out the video push method according to any one of claims 1 to 6.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911222506.6A CN110971973A (en) | 2019-12-03 | 2019-12-03 | Video pushing method and device and electronic equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911222506.6A CN110971973A (en) | 2019-12-03 | 2019-12-03 | Video pushing method and device and electronic equipment |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110971973A true CN110971973A (en) | 2020-04-07 |
Family
ID=70032770
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201911222506.6A Pending CN110971973A (en) | 2019-12-03 | 2019-12-03 | Video pushing method and device and electronic equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110971973A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111447239A (en) * | 2020-04-13 | 2020-07-24 | 北京字节跳动网络技术有限公司 | Video stream playing control method, device and storage medium |
| CN115577140A (en) * | 2022-09-16 | 2023-01-06 | 广州汽车集团股份有限公司 | Video recommendation method and device, vehicle and storage medium |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102473315A (en) * | 2009-08-20 | 2012-05-23 | 汤姆森特许公司 | Method and apparatus for reusing tree structures to encode and decode binary sets |
| CN103686237A (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 resource |
| US20170024471A1 (en) * | 2015-07-24 | 2017-01-26 | Xiaomi Inc. | Method and device for information push, and storage medium |
| CN107832437A (en) * | 2017-11-16 | 2018-03-23 | 北京小米移动软件有限公司 | Audio/video method for pushing, device, equipment and storage medium |
| CN107888950A (en) * | 2017-11-09 | 2018-04-06 | 福州瑞芯微电子股份有限公司 | A kind of method and system for recommending video |
| CN109558514A (en) * | 2019-01-08 | 2019-04-02 | 青岛聚看云科技有限公司 | Video recommendation method, its device, information processing equipment and storage medium |
-
2019
- 2019-12-03 CN CN201911222506.6A patent/CN110971973A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102473315A (en) * | 2009-08-20 | 2012-05-23 | 汤姆森特许公司 | Method and apparatus for reusing tree structures to encode and decode binary sets |
| CN103686237A (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 resource |
| US20170024471A1 (en) * | 2015-07-24 | 2017-01-26 | Xiaomi Inc. | Method and device for information push, and storage medium |
| CN107888950A (en) * | 2017-11-09 | 2018-04-06 | 福州瑞芯微电子股份有限公司 | A kind of method and system for recommending video |
| CN107832437A (en) * | 2017-11-16 | 2018-03-23 | 北京小米移动软件有限公司 | Audio/video method for pushing, device, equipment and storage medium |
| CN109558514A (en) * | 2019-01-08 | 2019-04-02 | 青岛聚看云科技有限公司 | Video recommendation method, its device, information processing equipment and storage medium |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111447239A (en) * | 2020-04-13 | 2020-07-24 | 北京字节跳动网络技术有限公司 | Video stream playing control method, device and storage medium |
| WO2021208607A1 (en) * | 2020-04-13 | 2021-10-21 | 北京字节跳动网络技术有限公司 | Video stream playing control method and apparatus, and storage medium |
| US11889135B2 (en) | 2020-04-13 | 2024-01-30 | Douyin Vision Co., Ltd. | Video stream playing control method and apparatus, and storage medium |
| CN115577140A (en) * | 2022-09-16 | 2023-01-06 | 广州汽车集团股份有限公司 | Video recommendation method and device, vehicle and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109460513B (en) | Method and apparatus for generating click rate prediction model | |
| TWI702844B (en) | Method, device, apparatus, and storage medium of generating features of user | |
| CN109819284B (en) | Short video recommendation method and device, computer equipment and storage medium | |
| CN106326391B (en) | Multimedia resource recommendation method and device | |
| US20180336193A1 (en) | Artificial Intelligence Based Method and Apparatus for Generating Article | |
| CN110909182B (en) | Multimedia resource searching method, device, computer equipment and storage medium | |
| CN110543598B (en) | Information recommendation method and device and terminal | |
| CN110991187B (en) | Entity linking method, device, electronic equipment and medium | |
| CN111178970B (en) | Advertisement putting method and device, electronic equipment and computer readable storage medium | |
| WO2020088058A1 (en) | Information generating method and device | |
| CN109511015B (en) | Multimedia resource recommendation method, device, storage medium and equipment | |
| CN111708901A (en) | Multimedia resource recommendation method and device, electronic equipment and storage medium | |
| CN114707074B (en) | Content recommendation method, device and system | |
| CN111461841A (en) | Article recommendation method, device, server and storage medium | |
| CN108197336B (en) | Video searching method and device | |
| CN111597446B (en) | Content pushing method and device based on artificial intelligence, server and storage medium | |
| CN113656681B (en) | Object evaluation method, device, equipment and storage medium | |
| CN107977678A (en) | Method and apparatus for output information | |
| CN111883131B (en) | Voice data processing method and device | |
| CN115618024A (en) | Multimedia recommendation method and device and electronic equipment | |
| CN110971973A (en) | Video pushing method and device and electronic equipment | |
| CN115687690A (en) | Video recommendation method and device, electronic equipment and storage medium | |
| CN111695041A (en) | Method and device for recommending information | |
| CN111753126A (en) | Method and device for video dubbing | |
| CN108647986B (en) | Target user determination method and device and electronic equipment |
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 | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20200407 |
|
| RJ01 | Rejection of invention patent application after publication |