CN112256969A - Media material pushing method and device, electronic equipment and storage medium - Google Patents

Media material pushing method and device, electronic equipment and storage medium Download PDF

Info

Publication number
CN112256969A
CN112256969A CN202011171038.7A CN202011171038A CN112256969A CN 112256969 A CN112256969 A CN 112256969A CN 202011171038 A CN202011171038 A CN 202011171038A CN 112256969 A CN112256969 A CN 112256969A
Authority
CN
China
Prior art keywords
media material
media
materials
sample
candidate
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.)
Granted
Application number
CN202011171038.7A
Other languages
Chinese (zh)
Other versions
CN112256969B (en
Inventor
董琦
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Beijing ByteDance Network Technology Co Ltd
Original Assignee
Beijing ByteDance Network Technology Co Ltd
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Beijing ByteDance Network Technology Co Ltd filed Critical Beijing ByteDance Network Technology Co Ltd
Priority to CN202011171038.7A priority Critical patent/CN112256969B/en
Publication of CN112256969A publication Critical patent/CN112256969A/en
Application granted granted Critical
Publication of CN112256969B publication Critical patent/CN112256969B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/90—Details of database functions independent of the retrieved data types
    • G06F16/95—Retrieval from the web
    • G06F16/953—Querying, e.g. by the use of web search engines
    • G06F16/9535—Search customisation based on user profiles and personalisation
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/40—Information retrieval; Database structures therefor; File system structures therefor of multimedia data, e.g. slideshows comprising image and additional audio data
    • G06F16/43—Querying
    • G06F16/435—Filtering based on additional data, e.g. user or group profiles
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/40—Information retrieval; Database structures therefor; File system structures therefor of multimedia data, e.g. slideshows comprising image and additional audio data
    • G06F16/45—Clustering; Classification
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/90—Details of database functions independent of the retrieved data types
    • G06F16/95—Retrieval from the web
    • G06F16/958—Organisation or management of web site content, e.g. publishing, maintaining pages or automatic linking
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L67/00—Network arrangements or protocols for supporting network services or applications
    • H04L67/50—Network services
    • H04L67/55—Push-based network services

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Databases & Information Systems (AREA)
  • Data Mining & Analysis (AREA)
  • Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Multimedia (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Signal Processing (AREA)
  • Electrically Operated Instructional Devices (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)

Abstract

The present disclosure provides a media material pushing method, a media material pushing device, an electronic device, and a storage medium, wherein the pushing method includes: selecting a first media material meeting a preset condition based on user behavior data corresponding to the pushed media material; selecting a target media material matched with the first media material from candidate media materials to be pushed based on the first media material and user operation characteristics corresponding to each second media material in a media material library; and pushing the target media material.

Description

Media material pushing method and device, electronic equipment and storage medium
Technical Field
The present disclosure relates to the field of computer technologies, and in particular, to a method and an apparatus for pushing a media material, an electronic device, and a storage medium.
Background
With the development of internet technology, various applications (apps) are gradually emerging, and a large amount of media materials can be popularized for users through the applications, for example, for a media Application, graphics media materials, video media materials, and the like can be popularized for users.
When media materials are popularized to users, the latest media materials are generally pushed to the users according to the release time of the media materials; or, personalized media materials are directly pushed to the user according to the user preference. The media material pushed in the above manner may not attract users, resulting in a low effective utilization of the media material.
Disclosure of Invention
The embodiment of the disclosure provides at least one media material pushing scheme, so as to improve the effective utilization rate of the pushed media materials.
In a first aspect, an embodiment of the present disclosure provides a method for pushing a media material, including:
selecting a first media material meeting a preset condition based on user behavior data corresponding to the pushed media material;
selecting a target media material matched with the first media material from candidate media materials to be pushed based on the first media material and user operation characteristics corresponding to each second media material in a media material library;
and pushing the target media material.
In a possible implementation manner, the pushed media materials are media materials which are pushed by other application programs except the target application program and whose release sources are the target application program;
selecting a target media material matched with the first media material from candidate media materials to be pushed based on the first media material and the user operation characteristics corresponding to the second media materials in the media material library, wherein the selecting comprises the following steps:
selecting a target media material matched with the first media material from candidate media materials to be pushed corresponding to the target application program based on the user operation characteristics corresponding to the first media material and each second media material when the first media material and each second media material are displayed at a client side associated with the target application program;
the pushing the target media material comprises:
and pushing the target media material through other application programs except the target application program.
In one possible embodiment, before selecting the target media material matching the first media material from the candidate media materials to be pushed, the pushing method further comprises:
determining a target time period for selecting the candidate media material based on the release time information corresponding to the first media material;
and selecting the candidate media materials from the media material library based on the corresponding user operation characteristics of the second media materials in the target time period.
In one possible embodiment, the selecting, from the candidate media materials to be pushed, a target media material that matches the first media material includes:
respectively determining a first media material vector corresponding to the first media material and a candidate media material vector corresponding to each candidate media material according to a weight matrix of a pre-trained neural network; the weight matrix of the pre-trained neural network is obtained by training based on the user operation characteristics, the first media material and the candidate media material;
and determining a target media material matched with the first media material based on the first media material vector and the candidate media material vector corresponding to each candidate media material.
In one possible implementation, the determining, based on the first media material vector and candidate media material vectors corresponding to candidate media materials, a target media material that matches the first media material includes:
determining similarity between the first media material and each candidate media material based on the first media material vector and the candidate media material vector corresponding to each candidate media material;
according to the similarity between the first media material and each candidate media material, performing descending sorting on the candidate media materials corresponding to the first media material;
and selecting candidate media materials with the similarity exceeding a preset similarity threshold value and with a set number before sorting as the target media materials.
In one possible embodiment, the weight matrix of the neural network is trained based on the user-operated features, the first media material, and the candidate media materials in the following manner:
forming a sample material library based on the first media material and the candidate media materials, wherein the sample material library comprises a plurality of sample materials;
determining positive sample materials and negative sample materials contained in the sample material library based on user operation characteristics corresponding to the sample materials;
training a neural network based on the positive sample material and the negative sample material respectively, and obtaining a weight matrix corresponding to the neural network after the neural network reaches a preset convergence condition.
In a possible implementation manner, the determining, based on the user operation characteristics corresponding to each sample material, positive sample materials and negative sample materials contained in the sample material library includes:
if different sample materials have the same preset user operation characteristics for the same user, determining that the different sample materials form a group of positive sample materials;
and if the different sample materials do not have the same preset user operation characteristics aiming at the same user, determining that the different sample materials form a group of negative sample materials.
In one possible embodiment, the determining the positive sample material and the negative sample material included in the sample material library based on the user operation characteristics corresponding to the sample materials includes:
if different sample materials exist, the feature vectors formed by the same user aiming at the user operation features of all the sample materials in the different sample materials are extracted;
determining whether the similarity of the user operation characteristics of the same user for each sample material in the different sample materials is greater than a set similarity threshold value or not based on a feature vector formed by the user operation characteristics respectively corresponding to the same user;
and if the similarity of the user operation characteristics of the same user for each sample material in the different sample materials is greater than the set similarity threshold, determining that the different sample materials form a positive sample material, otherwise, determining that the different sample materials form a negative sample material.
In a second aspect, an embodiment of the present disclosure provides a media material pushing apparatus, including:
the first selection module is used for selecting a first media material meeting a preset condition based on user behavior data corresponding to the pushed media material;
the second selection module is used for selecting a target media material matched with the first media material from candidate media materials to be pushed based on the first media material and the user operation characteristics corresponding to each second media material in the media material library;
and the material pushing module is used for pushing the target media material.
In a possible implementation manner, the pushed media materials are media materials which are pushed by other application programs except the target application program and whose release sources are the target application program;
the second selection module, when configured to select a target media material matched with the first media material from candidate media materials to be pushed based on the first media material and user operation features corresponding to second media materials in the media material library, includes:
selecting a target media material matched with the first media material from candidate media materials to be pushed corresponding to the target application program based on the user operation characteristics corresponding to the first media material and each second media material when the first media material and each second media material are displayed at a client side associated with the target application program;
when the material pushing module is used for pushing the target media material, the material pushing module comprises:
and pushing the target media material through other application programs except the target application program.
In a possible implementation manner, before the second selection module selects the target media material matching the first media material from the candidate media materials to be pushed, the second selection module is further configured to:
determining a target time period for selecting the candidate media material based on the release time information corresponding to the first media material;
and selecting the candidate media materials from the media material library based on the corresponding user operation characteristics of the second media materials in the target time period.
In one possible embodiment, the second selection module, when configured to select a target media material matching the first media material from among the candidate media materials to be pushed, comprises:
respectively determining a first media material vector corresponding to the first media material and a candidate media material vector corresponding to each candidate media material according to a weight matrix of a pre-trained neural network; the weight matrix of the pre-trained neural network is obtained by training based on the user operation characteristics, the first media material and the candidate media material;
and determining a target media material matched with the first media material according to the first media material vector and the candidate media material vector corresponding to each candidate media material.
In one possible implementation, the second selection module, when configured to determine, based on the first media material and candidate media material vectors corresponding to the candidate media materials, a target media material matching the first media material, includes:
determining similarity between the first media material and each candidate media material based on the first media material vector and the candidate media material vector corresponding to each candidate media material;
according to the similarity between the first media material and each candidate media material, performing descending sorting on the candidate media materials corresponding to the first media material;
and selecting candidate media materials with the similarity exceeding a preset similarity threshold value and with a set number before sorting as the target media materials.
In one possible implementation, the push device further includes a network training module for training a weight matrix of the neural network based on the user operation features, the first media material, and the candidate media material in the following manner:
forming a sample material library based on the first media material and the candidate media materials, wherein the sample material library comprises a plurality of sample materials;
determining positive sample materials and negative sample materials contained in the sample material library based on user operation characteristics corresponding to the sample materials;
and training the neural network based on the positive sample material and the negative sample material respectively, and obtaining a weight matrix corresponding to the neural network after the neural network reaches a preset convergence condition.
In a possible implementation manner, when the network training module is configured to determine, based on a user operation feature corresponding to each sample material, positive sample materials and negative sample materials contained in the sample material library, the network training module includes:
if different sample materials have the same preset user operation characteristics for the same user, determining that the different sample materials form a group of positive sample materials;
and if the different sample materials do not have the same preset user operation characteristics aiming at the same user, determining that the different sample materials form a group of negative sample materials.
In a possible implementation manner, the user operation characteristics include a plurality of types, and the network training module, when configured to determine, based on the user operation characteristics corresponding to each sample material, positive sample materials and negative sample materials included in the sample material library, includes:
if different sample materials exist, the feature vectors formed by the same user aiming at the user operation features of all the sample materials in the different sample materials are extracted;
determining whether the similarity of the user operation characteristics of the same user for each sample material in the different sample materials is greater than a set similarity threshold value or not based on a feature vector formed by the user operation characteristics respectively corresponding to the same user;
and if the similarity of the user operation characteristics of the same user for each sample material in the different sample materials is greater than the set similarity threshold, determining that the different sample materials form a positive sample material, otherwise, determining that the different sample materials form a negative sample material.
In a third aspect, an embodiment of the present disclosure provides an electronic device, including: a processor, a memory and a bus, the memory storing machine-readable instructions executable by the processor, the processor and the memory communicating over the bus when the electronic device is running, the machine-readable instructions when executed by the processor performing the steps of the push method according to the first aspect.
In a fourth aspect, the disclosed embodiments provide a computer-readable storage medium having stored thereon a computer program, which, when executed by a processor, performs the steps of the push method according to the first aspect.
According to the method for pushing the media materials, first, according to user behavior data corresponding to the pushed media materials, a first media material meeting preset conditions is selected, for example, the first media material with high effective utilization rate can be selected, and further, based on user operation characteristics corresponding to the first media material and second media materials in a media material library, a target media material matched with the first media material is selected from candidate media materials to be pushed.
In order to make the aforementioned objects, features and advantages of the present disclosure more comprehensible, preferred embodiments accompanied with figures are described in detail below.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present disclosure, the drawings required for use in the embodiments will be briefly described below, and the drawings herein incorporated in and forming a part of the specification illustrate embodiments consistent with the present disclosure and, together with the description, serve to explain the technical solutions of the present disclosure. It is appreciated that the following drawings depict only certain embodiments of the disclosure and are therefore not to be considered limiting of its scope, for those skilled in the art will be able to derive additional related drawings therefrom without the benefit of the inventive faculty.
Fig. 1 shows a flowchart of a method for pushing media materials according to an embodiment of the present disclosure;
FIG. 2 is a flow chart illustrating a method of determining targeted media material provided by an embodiment of the present disclosure;
FIG. 3 is a flow chart of a method for specifically targeting media material provided by an embodiment of the present disclosure;
FIG. 4 illustrates a flow chart of a method of determining a weight matrix for a neural network provided by an embodiment of the present disclosure;
fig. 5 is a schematic diagram illustrating a network structure of a neural network provided by an embodiment of the present disclosure;
fig. 6 is a schematic structural diagram of a media material pushing device provided by an embodiment of the present disclosure;
fig. 7 shows a schematic diagram of an electronic device provided by an embodiment of the present disclosure.
Detailed Description
In order to make the objects, technical solutions and advantages of the embodiments of the present disclosure more clear, the technical solutions of the embodiments of the present disclosure will be described clearly and completely with reference to the drawings in the embodiments of the present disclosure, and it is obvious that the described embodiments are only a part of the embodiments of the present disclosure, not all of the embodiments. The components of the embodiments of the present disclosure, generally described and illustrated in the figures herein, can be arranged and designed in a wide variety of different configurations. Thus, the following detailed description of the embodiments of the present disclosure, presented in the figures, is not intended to limit the scope of the claimed disclosure, but is merely representative of selected embodiments of the disclosure. All other embodiments, which can be derived by a person skilled in the art from the embodiments of the disclosure without making creative efforts, shall fall within the protection scope of the disclosure.
It should be noted that: like reference numbers and letters refer to like items in the following figures, and thus, once an item is defined in one figure, it need not be further defined and explained in subsequent figures.
The term "and/or" herein merely describes an associative relationship, meaning that three relationships may exist, e.g., a and/or B, may mean: a exists alone, A and B exist simultaneously, and B exists alone. In addition, the term "at least one" herein means any one of a plurality or any combination of at least two of a plurality, for example, including at least one of A, B, C, and may mean including any one or more elements selected from the group consisting of A, B and C.
With the development of internet technology, various application programs are gradually developed, a large number of media materials can be popularized for users through the application programs, and when media materials are pushed to users, how to select media materials attracting users from a large number of candidate media materials is achieved, so that the effective utilization rate of media resources is improved, and the technical problem to be solved is urgently needed.
Based on the above research, the method for pushing media materials provided in the embodiment of the present disclosure may select, according to user behavior data corresponding to a pushed media material, a first media material that meets a preset condition, for example, a first media material with a higher effective utilization rate, and further select, based on user operation characteristics corresponding to the first media material and second media materials in a media material library, a target media material that matches the first media material from candidate media materials to be pushed, because media materials liked by the same user have certain similarity, a target media material that has similar characteristics to the first media material may be selected from the candidate media materials in such a manner, so that when the target media material is pushed, the effective utilization rate of the pushed target media material may be improved.
To facilitate understanding of the present embodiment, first, a detailed description is given to a pushing method of a media material disclosed in an embodiment of the present disclosure, where an execution subject of the pushing method of the media material provided in an embodiment of the present disclosure is generally a computer device with certain computing capability, and the computer device includes, for example: terminal equipment or servers or other processing devices. In some possible implementations, the method of pushing media material may be implemented by a processor calling computer readable instructions stored in a memory.
Referring to fig. 1, a flowchart of a method for pushing media materials according to an embodiment of the present disclosure is shown, where the method for pushing media materials includes the following steps S101 to S103:
s101, selecting a first media material meeting a preset condition based on user behavior data corresponding to the pushed media material.
For example, the media material may be published by the target application and then pushed to other applications except the target application for browsing and downloading the target application by users of the other applications, for example, the target application is media application a, and the other applications are other media applications except the media application a.
In order to select the first media material that satisfies the preset condition in the latest time period in consideration of the difference in time period and the media materials preferred by the user, the first media material that satisfies the preset condition may be selected from among the media materials that have been pushed in the set history time period, for example, the first media material may be selected from among the media materials that have been pushed in the latest half year.
For example, the user behavior data corresponding to the pushed media material is user behavior data authorized by the user, and may specifically include behavior data such as a click of the media material by the user when the media material is displayed in another application program, and a download of the media material for a target application program after the click of the media material.
By counting the user behavior data corresponding to each media material in a set time period, at least one of a click rate corresponding to the media material, a download amount of the target application triggered by the media material and a conversion rate triggered by the media material can be determined, wherein the conversion rate is a ratio of the number of times of triggering downloading the target application after clicking the media material to the number of times of clicking the media material, and then a first media material can be further selected based on one or more of the click rate, the download amount and the conversion rate, for example, a media material with a click rate greater than a set click rate threshold value is selected as the first media material, or a media material with a download amount greater than a set download amount threshold value is selected as the first media material, or a media material with a conversion rate greater than a set conversion rate threshold value is selected as the first media material, or selecting the media material with the click rate larger than the set click rate threshold, the download amount larger than the set download amount threshold and the conversion rate larger than the set conversion rate threshold as the first media material, and selecting the first media material popular in the user group according to the mode.
And S102, selecting a target media material matched with the first media material from candidate media materials to be pushed based on the first media material and the user operation characteristics corresponding to the second media materials in the media material library.
Exemplarily, taking the above mentioned pushed media materials as media materials whose distribution sources are pushed by other applications besides the target application as target applications, when selecting the target media materials matching the first media materials from the candidate media materials to be pushed based on the user operation characteristics corresponding to the first media materials and the second media materials in the media material library, the method includes:
and selecting the target media material matched with the first media material from candidate media materials to be pushed corresponding to the target application program based on the user operation characteristics corresponding to the first media material and the second media materials when the first media material and the second media materials are displayed at the client side associated with the target application program.
For example, the target application may be an application to be promoted, for example, if it is desired to promote a news application, the target application may be the news application.
The media material of which the release source is the target application program refers to the media material stored and released by the server corresponding to the target application program.
Illustratively, the media material library may include a large amount of second media materials pre-stored in a server corresponding to the target application program, and the second media materials or the first media materials may correspond to user operation characteristics of at least one user when displayed on a client associated with the target application program, where the user operation characteristics of the user for the media materials may include operation characteristics of the user for marking like, collection, comment, forwarding, sharing, downloading, and the like of the media materials, and the user operation characteristics are user operation characteristics authorized by the user.
Considering that the media material library contains a large amount of second media materials, before selecting the target media material matched with the first media material, the candidate media materials to be pushed can be selected, for example, the candidate media materials to be pushed corresponding to the target application program can be screened out, so that the range of selecting the target media material can be narrowed, and the efficiency of determining the target media material can be improved.
For example, when selecting the candidate media materials in the media material library, the second media materials with higher operation times may be selected as the candidate media materials based on the user operation characteristics corresponding to the second media materials, for example, the second media materials that are preferred by the user marks exceeding the set number may be selected as the candidate media materials, or the second media materials that are clicked and collected by the users exceeding the set number may be selected as the candidate media materials, or the second media materials that meet the set conditions may also be selected as the candidate media materials by one or more other user operation characteristics, which is not specifically limited herein.
Here, when the target media material is selected, the target media material is selected by the user operation characteristics corresponding to the first media material and each second media material in the media material library, and for example, when the first media material is a media material that is liked by young women, the target media material that is selected by the user operation characteristics and matches the first media material is likely to be liked by young women when the target media material is pushed.
And S103, pushing the target media material.
For example, after the target media material is obtained, the target media material may be pushed by other applications besides the target application, for example, the target application is a video-class application, the target media material is a video a, a video B, and a video C played through the video-class application, the video a, the video B, and the video C may be played through other applications, and when a user of the other applications clicks on the video a, the video B, or the video C for viewing, the user may jump to the video-class application and play through the video-class application, so that the user experiences and uses the video-class application.
According to the method for pushing the media materials, first, according to user behavior data corresponding to the pushed media materials, a first media material meeting preset conditions is selected, for example, the first media material with high effective utilization rate can be selected, and further, based on user operation characteristics corresponding to the first media material and second media materials in a media material library, a target media material matched with the first media material is selected from candidate media materials to be pushed.
For the above S101, when selecting the first media material that satisfies the preset condition, the method may include;
selecting at least one media material meeting preset conditions from the pushed media materials, wherein the at least one media material meets the preset conditions and is selected from the download amount of a media material trigger target application program, the click rate of the media material and the conversion rate triggered by the media material; the conversion rate is the ratio of the number of times of triggering downloading of the target application program after the media material is clicked to the number of times of clicking the media material.
For explaining the selection process of the first media material, a return result corresponding to the media material may be introduced, for example, the return result corresponding to the pushed media material may include at least one of a click rate corresponding to the media material, a download amount of the target application triggered by the media material, and a conversion rate triggered by the media material, when the first media material meeting the preset condition is selected, the media material whose return result meets the preset condition may be selected as the first media material, for example, the first media material whose at least one of the download amount of the target application, the click rate of the media material, and the conversion rate triggered by the media material in the return result meets the preset condition is selected.
The click rate of the media material can be determined by the ratio of the click times and the display times of the media material.
Illustratively, when the reward result corresponding to the media material simultaneously includes the download amount for the target application program, the click rate for the media material and the conversion rate triggered by the media material, the reward result meeting the preset condition includes at least one or more of the following:
(1) the downloading amount is greater than a preset downloading amount threshold value, the click rate is greater than a preset click rate threshold value, and the conversion rate is greater than a preset conversion rate threshold value;
(2) and after the download amount, the click rate and the conversion rate are subjected to weighted summation, the obtained weighted summation result is greater than a preset value.
Exemplarily, assuming that a preset downloading amount threshold is m, a preset click rate threshold is n, a conversion rate is k, and a preset score is t; taking the first case as an example, if the download amount of a media material is greater than m, the click rate is greater than n, and the conversion rate is greater than k when the media material is pushed by other application programs, the media material can be used as a first media material; taking the second case as an example, after weighted summation is performed on the download amount, the click rate and the conversion rate corresponding to one piece of media material, the obtained weighted summation result is greater than t, and the media material can also be used as the first media material; when the download amount, the click rate and the conversion rate are weighted and summed, the weights corresponding to the download amount, the click rate and the conversion rate can be set in advance based on big data statistics; taking the first case and the second case that are satisfied at the same time, if a media material is pushed by other application programs, the corresponding return result is that the download amount is greater than m, the click rate is greater than n, the conversion rate is greater than k, and after the download amount, the click rate and the conversion rate corresponding to the media material are weighted and summed, the obtained weighted sum result is greater than t, the media material can be used as the first media material.
In the embodiment of the disclosure, by selecting the first media material of which the return result meets the preset condition, the media material with higher effective utilization rate can be selected from the pushed media materials as the standard for screening other target media materials to be pushed, and based on the standard, the media materials with the same effective utilization rate can be screened.
In one embodiment, before selecting a target media material matching the first media material from candidate media materials to be pushed, the pushing method provided by the embodiment of the disclosure further includes:
(1) determining a target time period for selecting candidate media materials based on the release time information corresponding to the first media material;
(2) and selecting candidate media materials from the media material library based on the corresponding user operation characteristics of the second media materials in the target time period.
Considering that the media material library includes a huge amount of second media materials, in order to quickly extract candidate media materials comparable to the first media materials from the huge amount of second media material library, that is, to make the user operation features corresponding to the candidate media materials comparable to the user operation features corresponding to the first media materials, the target time period for selecting the candidate media materials may be determined based on the distribution time information of the first media materials in the target application program, specifically, for the distribution time information corresponding to the first media materials, a time period between the earliest distribution time and the latest distribution time of the first media materials may be taken as the target time period, for example, the first media materials include 3 pieces, which are respectively marked as a first media material a, a first media material B and a first media material C, and the distribution time in the target application program of the first media material a is 2020, 4 months and 1 days, when the distribution time of the first media material B in the target application is 4/5/2020 and the distribution time of the first media material C in the target application is 4/10/2020, 1/2020 to 4/10/2020 can be used as the target time period.
After the target time period is determined, candidate media materials may be selected from the media material library based on user operation characteristics corresponding to each second media material in the media material library in the target time period, specifically, for a case that the user operation characteristics mentioned above include consumption characteristics such as mark like, collection, comment, forwarding, sharing, and downloading, when selecting the candidate media materials, a second media material in which at least one user operation characteristic satisfies a preset condition may be selected as the candidate media material, for example, a second media material in which the number of times of mark like exceeds a set number in the target time period in the media material library is selected as the candidate media material.
After selecting candidate media materials from the media material library, when selecting a target media material matching the first media material from the candidate media materials to be pushed, as shown in fig. 2, the following steps S201 to S202 may be included:
s201, respectively determining a first media material vector corresponding to a first media material and a candidate media material vector corresponding to each candidate media material according to a pre-trained weight matrix of a neural network; the weight matrix of the pre-trained neural network is obtained by training based on user operation characteristics, the first media material and the candidate media material;
s202, determining target media materials matched with the first media materials based on the first media material vectors and the candidate media material vectors corresponding to the candidate media materials.
Exemplarily, the weight matrix may map the first media material to a corresponding first media material vector, and map the candidate media material to a corresponding candidate media material vector, specifically, a neural network capable of converting the media material into the media material vector may be trained through the user operation characteristics, the first media material, and the candidate media material, and a weight matrix corresponding to the neural network is obtained after the training is finished, and a process of specifically training the neural network to obtain the weight matrix will be described later.
According to the weight matrix of the pre-trained neural network, the first media material or the candidate media material can be converted into a corresponding media material vector, and the similarity between the first media material and the candidate media material can be obtained through calculation of the first media material vector and the candidate media material vector, for example, the cosine similarity between two media material vectors can be calculated, and the similarity between the first media material and the candidate media material can be represented through the cosine similarity.
In the embodiment of the disclosure, the target media material matched with the first media material is selected from the candidate media materials, so that the target media material with the same effective utilization rate as the first media material can be obtained.
Specifically, in S202, when determining a target media material matching the first media material based on the first media material vector and the candidate media material vectors corresponding to the candidate media materials, as shown in fig. 3, the following steps S301 to S303 may be included:
s301, determining similarity between the first media material and each candidate media material based on the first media material vector and the candidate media material vector corresponding to each candidate media material;
s302, according to the similarity between the first media material and each candidate media material, performing descending sorting on the candidate media materials corresponding to the first media material;
and S303, selecting candidate media materials with the similarity exceeding a preset similarity threshold and with the set number before sorting as target media materials.
When the target media material matched with the first media material is determined, the similarity between the first media material vector corresponding to the first media material and the candidate media material vectors corresponding to the candidate media materials can be determined according to a cosine similarity formula, so that the similarity between the first media material and the candidate media materials can be obtained, for example, when the candidate media materials include 1000 pieces, 1000 similarities corresponding to the first media material can be obtained for the first piece of the first media material.
And for the 1000 similarities corresponding to the first media material, sorting the 1000 candidate media materials in a descending order according to a decreasing similarity mode, and further selecting the candidate media materials with the similarity exceeding a preset similarity threshold value from the 1000 candidate media materials, wherein the candidate media materials with the preset number before sorting are used as target media materials matched with the first media material.
In the case where the first media material includes a plurality of pieces, the target media material matching each first media material can be obtained in the same manner.
Specifically, for the weight matrix of the above-mentioned neural network, as shown in fig. 4, the weight matrix may be obtained by training the neural network based on the user operation features, the first media material and the candidate media material in the following manner, specifically including the following S401 to S403:
s401, forming a sample material library based on the first media material and the candidate media materials, wherein the sample material library comprises a plurality of sample materials;
s402, determining positive sample materials and negative sample materials contained in a sample material library based on user operation characteristics corresponding to the sample materials;
and S403, training the neural network based on the positive sample material and the negative sample material respectively, and obtaining a weight matrix corresponding to the neural network after the neural network reaches a preset convergence condition.
When the weight matrix is obtained by training the neural network, according to the determination method of the first media material and the candidate media material mentioned above in the present disclosure, after the first media material and the candidate media material are obtained, the first media material and the candidate media material may form a sample material library, and each first media material or candidate media material is used as a sample material in the sample material library.
Before training the neural network, sample materials in a sample material library are classified, for example, the sample materials can be divided into multiple groups of positive sample materials and multiple groups of negative sample materials according to user operation characteristics corresponding to the sample materials, and each group of positive sample materials and each group of negative sample materials comprise two sample materials.
In one embodiment, when determining the positive sample material and the negative sample material contained in the sample material library based on the user operation characteristics corresponding to each sample material, the determining may include:
(1) if different sample materials have the same preset user operation characteristics for the same user, determining that the different sample materials form a group of positive sample materials;
(2) and if the different sample materials do not have the same preset user operation characteristics aiming at the same user, determining that the different sample materials form a group of negative sample materials.
For example, when determining the positive sample materials and the negative sample materials, the above determination may be performed in a manner of every two sample materials, so as to obtain multiple sets of positive sample materials and multiple sets of negative sample materials.
The preset user operation characteristic may be one or more specific user operation characteristics, for example, the user operation characteristic "mark like" mentioned above may be used to mark a sample material that is liked by the same user as a positive sample material, and mark a sample material that is not liked by the same user as a negative sample material.
In another embodiment, the user operation characteristics include a plurality of types, and when determining the positive sample material and the negative sample material included in the sample material library based on the user operation characteristics corresponding to each sample material, the method may include:
(1) if different sample materials are operated by the same user, extracting feature vectors formed by the same user aiming at the user operation features of the sample materials in the different sample materials;
(2) determining whether the similarity of the user operation characteristics of the same user for each sample material in different sample materials is greater than a set similarity threshold value or not based on a feature vector formed by the user operation characteristics respectively corresponding to the same user;
(3) and if the similarity of the user operation characteristics of the same user for each sample material in different sample materials is greater than a set similarity threshold value, determining that the different sample materials form a positive sample material, otherwise, determining that the different sample materials form a negative sample material.
Illustratively, two sample materials operated by the same user, namely "zhang san", are recorded as sample material a and sample material B, respectively, if the user operation characteristics include: the user operation characteristics of the sample material A by the user Zhang III include likes, collections and comments, a feature vector formed by the user operation characteristics of the sample material A by the user Zhang III can be represented as [ 111000 ], the user operation characteristics of the sample material B by the user Zhang III include likes, collections, comments and forwards, and a feature vector formed by the user operation characteristics of the sample material B by the user Zhang III can be represented as [ 111100 ].
Furthermore, the similarity between the sample material A and the sample material B can be determined through the feature vectors [ 111000 ] and [ 111100 ] respectively corresponding to Zhang III of the user, and whether the similarity is greater than a set similarity threshold value is judged, for example, the similarity threshold value is set to be 0.8, whether the cosine similarity between the feature vector [ 111000 ] and the feature vector [ 111100 ] is greater than 0.8 can be judged, if so, the sample material A and the sample material B form a positive sample material, otherwise, the sample material A and the sample material B form a negative sample material.
In the embodiment of the disclosure, the positive sample material and the negative sample material contained in the sample material library can be judged according to whether the same user has operated the same, and based on the judgment, other media materials liked by the crowd who likes the first media material can be found, so that the target media material with the same effective utilization rate as the first media material can be obtained.
Particularly, the number of the negative sample materials determined in the two manners may be far greater than that of the positive sample materials, so that after the negative sample materials are obtained, the negative sample materials with the number close to that of the positive sample materials or the negative sample materials with the number slightly greater than that of the positive sample materials can be obtained through sampling, so as to meet the requirement on neural network training.
After the positive sample materials and the negative sample materials in the sample material library are obtained in the above manner, the neural network may be trained based on the positive sample materials and the negative sample materials, respectively, in the training process, in the embodiment of the present disclosure, based on the idea of word2vec, each sample material in the sample material library is initially encoded in the item2vec manner, so as to obtain one-hot codes corresponding to each sample material, for example, the sample material library includes 100 sample materials, the 100 sample materials are sorted and recorded as 1 to 100, the one-hot code corresponding to each sample material may be represented as a vector of 100 dimensions, the one-hot code corresponding to the 1 st sample material may be represented as [ 100.. 0], that is, the one-hot code corresponding to the 2 nd sample material may be represented as [ 010.. 0], namely, the one-hot codes corresponding to the sample materials in the sample material library can be obtained according to the method, wherein the second-bit eigenvalue is 1, and the rest eigenvalues are all vectors of 0.
Specifically, in the training process, a Skip-Gram network model structure may be introduced, as shown in fig. 5, a group of positive sample materials is formed by the sample material x and the sample material y1, the sample material y2, and the sample material y3, the sample material x and the sample material y1, the sample material y2, and the sample material y3 may be input into the network model shown in fig. 5, as shown in fig. 5, in the training process of the neural network, one-hot codes of the sample material x may be mapped to feature vectors by the weight matrix W, as [ h ] in fig. 51 ... hi ... hN-1 hN]Then, based on the eigenvector and the weight matrix W ', predictive coding vectors corresponding to the sample material y1, the sample material y2 and the sample material y3 are obtained, then loss values corresponding to the neural network are determined according to the one-hot coding and predictive coding vectors corresponding to the sample material y1, the sample material y2 and the sample material y3, the weight matrices W and W ' are adjusted continuously through the loss values, in addition, a negative sample material group is introduced to train the neural network, the loss values corresponding to the neural network are determined in the same way, the weight matrices W and W ' are adjusted through the loss values, according to the way, the neural network is trained continuously through the positive sample material and the negative sample material, and after the loss values are smaller than a set loss threshold or the training times reach a set time threshold (it is determined that the neural network reaches a preset convergence condition), obtaining a weight matrix of the trained neural network, wherein the weight matrix W 'enables a plurality of sample materials, which all form positive sample materials with the same sample materials, to be mapped into eigenvectors with higher similarity (for example, as shown in fig. 5, the similarity between eigenvectors determined by the sample materials y1, y2, and y3 through the weight matrix W' is higher), so that a plurality of sample materials, which all form negative sample materials with the same sample materials, are mapped into eigenvectors with lower similarity.
In the embodiment of the disclosure, a neural network is trained by judging positive sample materials and negative sample materials contained in a sample material library according to user operation characteristics, so as to obtain a weight matrix for mapping the sample materials into eigenvectors, and then target media materials similar to the first media materials can be determined based on the eigenvectors.
It will be understood by those skilled in the art that in the method of the present invention, the order of writing the steps does not imply a strict order of execution and any limitations on the implementation, and the specific order of execution of the steps should be determined by their function and possible inherent logic.
Based on the same technical concept, a push device corresponding to the push method of the media material is further provided in the embodiment of the present disclosure, and as the principle of solving the problem of the device in the embodiment of the present disclosure is similar to the push method in the embodiment of the present disclosure, the implementation of the device may refer to the implementation of the method, and repeated details are not described again.
Referring to fig. 6, there is shown a schematic diagram of a media material pushing apparatus 600 according to an embodiment of the present disclosure, the pushing apparatus includes:
a first selecting module 601, configured to select, based on user behavior data corresponding to a pushed media material, a first media material that meets a preset condition;
a second selecting module 602, configured to select, based on the first media material and user operation features corresponding to second media materials in the media material library, a target media material that matches the first media material from candidate media materials to be pushed;
the material pushing module 603 is configured to push the target media material.
In one embodiment, the pushed media materials are media materials which are pushed by other application programs except the target application program and whose release sources are the target application program;
the second selecting module 602, when configured to select a target media material matching the first media material from candidate media materials to be pushed based on the first media material and user operation features corresponding to the second media materials in the media material library, includes:
selecting a target media material matched with the first media material from candidate media materials to be pushed corresponding to a target application program based on corresponding user operation characteristics of the first media material and each second media material when the first media material and each second media material are displayed at a client side associated with the target application program;
the material pushing module 603, when used for pushing the target media material, includes:
the target media material is pushed by other applications than the target application.
In one embodiment, the second selection module 602 is further configured to, before selecting the target media material matching the first media material from the candidate media materials to be pushed:
determining a target time period for selecting candidate media materials based on the release time information corresponding to the first media material;
and selecting candidate media materials from the media material library based on the corresponding user operation characteristics of the second media materials in the target time period.
In one embodiment, the second selection module 602, when used to select a target media material matching the first media material from the candidate media materials to be pushed, comprises:
respectively determining a first media material vector corresponding to a first media material and a candidate media material vector corresponding to each candidate media material according to a pre-trained weight matrix of a neural network; the weight matrix of the pre-trained neural network is obtained based on user operation characteristics, the first media material and the candidate media material;
and determining a target media material matched with the first media material according to the first media material vector and the candidate media material vector corresponding to each candidate media material.
In one embodiment, the second selection module 602, when configured to determine the target media material matching the first media material based on the first media material and the candidate media material vector corresponding to each candidate media material, includes:
determining the similarity between the first media material and each candidate media material based on the first media material vector and the candidate media material vector corresponding to each candidate media material;
according to the similarity between the first media material and each candidate media material, performing descending sorting on the candidate media materials corresponding to the first media material;
and selecting candidate media materials with the similarity exceeding a preset similarity threshold value and with a set number before sorting as target media materials.
In one embodiment, the push device further comprises a network training module 604, the network training module 604 is configured to train a weight matrix of the neural network based on the user operation characteristics, the first media material, and the candidate media material in the following manner:
forming a sample material library based on the first media material and the candidate media material, wherein the sample material library comprises a plurality of sample materials;
determining positive sample materials and negative sample materials contained in a sample material library based on user operation characteristics corresponding to the sample materials;
and training the neural network based on the positive sample material and the negative sample material respectively, and obtaining a weight matrix corresponding to the neural network after the neural network reaches a preset convergence condition.
In one embodiment, the network training module 604 is configured to, when determining the positive sample materials and the negative sample materials contained in the sample material library based on the user operation characteristics corresponding to each sample material, include:
if different sample materials have the same preset user operation characteristics for the same user, determining that the different sample materials form a group of positive sample materials;
and if the different sample materials do not have the same preset user operation characteristics aiming at the same user, determining that the different sample materials form a group of negative sample materials.
In one embodiment, the user operation characteristics include a plurality of types, and the network training module 604, when configured to determine the positive sample materials and the negative sample materials included in the sample material library based on the user operation characteristics corresponding to each sample material, includes:
if different sample materials are operated by the same user, extracting feature vectors formed by the same user aiming at the user operation features of the sample materials in the different sample materials;
determining whether the similarity of the user operation characteristics of the same user for each sample material in different sample materials is greater than a set similarity threshold value or not based on a feature vector formed by the user operation characteristics respectively corresponding to the same user;
and if the similarity of the user operation characteristics of the same user for each sample material in different sample materials is greater than a set similarity threshold value, determining that the different sample materials form a positive sample material, otherwise, determining that the different sample materials form a negative sample material.
The description of the processing flow of each module in the device and the interaction flow between the modules may refer to the related description in the above method embodiments, and will not be described in detail here.
Corresponding to the method for pushing the media material in fig. 1, an embodiment of the present disclosure further provides an electronic device 700, as shown in fig. 7, a schematic structural diagram of the electronic device 700 provided in the embodiment of the present disclosure includes:
a processor 71, a memory 72, and a bus 73; the memory 72 is used for storing execution instructions and includes a memory 721 and an external memory 722; the memory 721 is also referred to as an internal memory, and is used for temporarily storing the operation data in the processor 71 and the data exchanged with the external memory 722 such as a hard disk, the processor 71 exchanges data with the external memory 722 through the memory 721, and when the electronic device 700 operates, the processor 71 communicates with the memory 72 through the bus 73, so that the processor 71 executes the following instructions: selecting a first media material meeting a preset condition based on user behavior data corresponding to the pushed media material; selecting a target media material matched with the first media material from candidate media materials to be pushed based on the first media material and user operation characteristics corresponding to each second media material in a media material library; and pushing the target media material.
The embodiments of the present disclosure also provide a computer-readable storage medium, on which a computer program is stored, where the computer program is executed by a processor to execute the steps of the method for pushing media materials in the above-mentioned method embodiments. The storage medium may be a volatile or non-volatile computer-readable storage medium.
The computer program product of the method for pushing a media material provided in the embodiments of the present disclosure includes a computer-readable storage medium storing a program code, where instructions included in the program code may be used to execute the steps of the method for pushing a media material described in the above method embodiments, which may be referred to in the above method embodiments specifically, and are not described herein again.
The embodiments of the present disclosure also provide a computer program, which when executed by a processor implements any one of the methods of the foregoing embodiments. The computer program product may be embodied in hardware, software or a combination thereof. In an alternative embodiment, the computer program product is embodied in a computer storage medium, and in another alternative embodiment, the computer program product is embodied in a Software product, such as a Software Development Kit (SDK), or the like.
It is clear to those skilled in the art that, for convenience and brevity of description, the specific working processes of the system and the apparatus described above may refer to the corresponding processes in the foregoing method embodiments, and are not described herein again. In the several embodiments provided in the present disclosure, it should be understood that the disclosed system, apparatus, and method may be implemented in other ways. The above-described embodiments of the apparatus are merely illustrative, and for example, the division of the units is only one logical division, and there may be other divisions when actually implemented, and for example, a plurality of units or components may be combined or integrated into another system, or some features may be omitted, or not executed. In addition, the shown or discussed mutual coupling or direct coupling or communication connection may be an indirect coupling or communication connection of devices or units through some communication interfaces, and may be in an electrical, mechanical or other form.
The units described as separate parts may or may not be physically separate, and parts displayed as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of the embodiment.
In addition, functional units in the embodiments of the present disclosure may be integrated into one processing unit, or each unit may exist alone physically, or two or more units are integrated into one unit.
The functions, if implemented in the form of software functional units and sold or used as a stand-alone product, may be stored in a non-volatile computer-readable storage medium executable by a processor. Based on such understanding, the technical solution of the present disclosure may be embodied in the form of a software product, which is stored in a storage medium and includes several instructions for causing a computer device (which may be a personal computer, a server, or a network device) to execute all or part of the steps of the method according to the embodiments of the present disclosure. And the aforementioned storage medium includes: various media capable of storing program codes, such as a usb disk, a removable hard disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk, or an optical disk.
Finally, it should be noted that: the above-mentioned embodiments are merely specific embodiments of the present disclosure, which are used for illustrating the technical solutions of the present disclosure and not for limiting the same, and the scope of the present disclosure is not limited thereto, and although the present disclosure is described in detail with reference to the foregoing embodiments, those skilled in the art should understand that: any person skilled in the art can modify or easily conceive of the technical solutions described in the foregoing embodiments or equivalent technical features thereof within the technical scope of the present disclosure; such modifications, changes or substitutions do not depart from the spirit and scope of the embodiments of the present disclosure, and should be construed as being included therein. Therefore, the protection scope of the present disclosure shall be subject to the protection scope of the claims.

Claims (11)

1. A method for pushing media material, comprising:
selecting a first media material meeting a preset condition based on user behavior data corresponding to the pushed media material;
selecting a target media material matched with the first media material from candidate media materials to be pushed based on the first media material and user operation characteristics corresponding to each second media material in a media material library;
and pushing the target media material.
2. A pushing method according to claim 1, wherein the pushed media material is media material pushed by an application other than the target application, the release source of which is the target application;
selecting a target media material matched with the first media material from candidate media materials to be pushed based on the first media material and the user operation characteristics corresponding to the second media materials in the media material library, wherein the selecting comprises the following steps:
selecting a target media material matched with the first media material from candidate media materials to be pushed corresponding to the target application program based on the user operation characteristics corresponding to the first media material and each second media material when the first media material and each second media material are displayed at a client side associated with the target application program;
the pushing the target media material comprises:
and pushing the target media material through other application programs except the target application program.
3. A push method according to claim 1 or 2, characterised in that before selecting a target media material matching the first media material from the candidate media materials to be pushed, the push method further comprises:
determining a target time period for selecting the candidate media material based on the release time information corresponding to the first media material;
and selecting the candidate media materials from the media material library based on the corresponding user operation characteristics of the second media materials in the target time period.
4. A push method according to claim 1 or 2, wherein said selecting a target media material from the candidate media materials to be pushed that matches the first media material comprises:
respectively determining a first media material vector corresponding to the first media material and a candidate media material vector corresponding to each candidate media material according to a weight matrix of a pre-trained neural network; the weight matrix of the pre-trained neural network is obtained by training based on the user operation characteristics, the first media material and the candidate media material;
and determining a target media material matched with the first media material based on the first media material vector and the candidate media material vector corresponding to each candidate media material.
5. The method of pushing according to claim 4, wherein determining a target media material matching the first media material based on the first vector of media material and a candidate vector of media material corresponding to each candidate media material comprises:
determining similarity between the first media material and each candidate media material based on the first media material vector and the candidate media material vector corresponding to each candidate media material;
according to the similarity between the first media material and each candidate media material, performing descending sorting on the candidate media materials corresponding to the first media material;
and selecting candidate media materials with the similarity exceeding a preset similarity threshold value and with a set number before sorting as the target media materials.
6. The push method of claim 4, wherein a weight matrix of the neural network is trained based on the user-operated features, the first media material, and the candidate media material in the following manner:
forming a sample material library based on the first media material and the candidate media materials, wherein the sample material library comprises a plurality of sample materials;
determining positive sample materials and negative sample materials contained in the sample material library based on user operation characteristics corresponding to the sample materials;
training a neural network based on the positive sample material and the negative sample material respectively, and obtaining a weight matrix corresponding to the neural network after the neural network reaches a preset convergence condition.
7. The method of claim 6, wherein the determining positive and negative sample materials contained in the sample material library based on the user operation characteristics corresponding to each sample material comprises:
if different sample materials have the same preset user operation characteristics for the same user, determining that the different sample materials form a group of positive sample materials;
and if the different sample materials do not have the same preset user operation characteristics aiming at the same user, determining that the different sample materials form a group of negative sample materials.
8. The method according to claim 6, wherein the user operation characteristics include a plurality of types, and the determining the positive sample material and the negative sample material included in the sample material library based on the user operation characteristics corresponding to each sample material includes:
if different sample materials exist, the feature vectors formed by the same user aiming at the user operation features of all the sample materials in the different sample materials are extracted;
determining whether the similarity of the user operation characteristics of the same user for each sample material in the different sample materials is greater than a set similarity threshold value or not based on a feature vector formed by the user operation characteristics respectively corresponding to the same user;
and if the similarity of the user operation characteristics of the same user for each sample material in the different sample materials is greater than the set similarity threshold, determining that the different sample materials form a positive sample material, otherwise, determining that the different sample materials form a negative sample material.
9. A device for pushing media material, comprising:
the first selection module is used for selecting a first media material meeting a preset condition based on user behavior data corresponding to the pushed media material;
the second selection module is used for selecting a target media material matched with the first media material from candidate media materials to be pushed based on the first media material and the user operation characteristics corresponding to each second media material in the media material library;
and the material pushing module is used for pushing the target media material.
10. An electronic device, comprising: processor, memory and bus, the memory storing machine-readable instructions executable by the processor, the processor and the memory communicating over the bus when the electronic device is running, the machine-readable instructions when executed by the processor performing the steps of the push method according to any of claims 1 to 8.
11. A computer-readable storage medium, having stored thereon a computer program which, when being executed by a processor, performs the steps of the push method according to any one of claims 1 to 8.
CN202011171038.7A 2020-10-28 2020-10-28 Media material pushing method and device, electronic equipment and storage medium Active CN112256969B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202011171038.7A CN112256969B (en) 2020-10-28 2020-10-28 Media material pushing method and device, electronic equipment and storage medium

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202011171038.7A CN112256969B (en) 2020-10-28 2020-10-28 Media material pushing method and device, electronic equipment and storage medium

Publications (2)

Publication Number Publication Date
CN112256969A true CN112256969A (en) 2021-01-22
CN112256969B CN112256969B (en) 2024-04-26

Family

ID=74261409

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202011171038.7A Active CN112256969B (en) 2020-10-28 2020-10-28 Media material pushing method and device, electronic equipment and storage medium

Country Status (1)

Country Link
CN (1) CN112256969B (en)

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103490981A (en) * 2013-09-11 2014-01-01 曹欢欢 Information pushing method and device across mobile applications
CN106960367A (en) * 2017-03-31 2017-07-18 北京猎豹移动科技有限公司 Promotion method, device and the server of application program
CN107657468A (en) * 2016-07-25 2018-02-02 北京金山云网络技术有限公司 Material evaluating method and device
CN107656947A (en) * 2016-10-24 2018-02-02 腾讯科技（北京）有限公司 A kind of multimedia messages player method and device
CN108959641A (en) * 2018-07-27 2018-12-07 北京未来媒体科技股份有限公司 A kind of content information recommended method and system based on artificial intelligence
CN111274884A (en) * 2020-01-11 2020-06-12 上海悠络客电子科技股份有限公司 Intelligent advertisement pushing system based on integration of face recognition and behavior recognition

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103490981A (en) * 2013-09-11 2014-01-01 曹欢欢 Information pushing method and device across mobile applications
CN107657468A (en) * 2016-07-25 2018-02-02 北京金山云网络技术有限公司 Material evaluating method and device
CN107656947A (en) * 2016-10-24 2018-02-02 腾讯科技（北京）有限公司 A kind of multimedia messages player method and device
CN106960367A (en) * 2017-03-31 2017-07-18 北京猎豹移动科技有限公司 Promotion method, device and the server of application program
CN108959641A (en) * 2018-07-27 2018-12-07 北京未来媒体科技股份有限公司 A kind of content information recommended method and system based on artificial intelligence
CN111274884A (en) * 2020-01-11 2020-06-12 上海悠络客电子科技股份有限公司 Intelligent advertisement pushing system based on integration of face recognition and behavior recognition

Also Published As

Publication number Publication date
CN112256969B (en) 2024-04-26

Similar Documents

Publication Publication Date Title
CN106503006B (en) Sequencing method and device for sub-applications in application App
CN106951571B (en) Method and device for labeling application with label
CN106294783A (en) A kind of video recommendation method and device
CN108304354B (en) Prediction model training method and device, storage medium and electronic equipment
CN110348907A (en) A kind of orientation method and device of advertisement crowd
CN106874503B (en) Method and device for acquiring recommended data
CN103984701A (en) Micro-blog forwarding quantity prediction model generation method and micro-blog forwarding quantity prediction method
CN108985048B (en) Simulator identification method and related device
CN111597446B (en) Content pushing method and device based on artificial intelligence, server and storage medium
CN113011471A (en) Social group dividing method, social group dividing system and related devices
CN111753204B (en) Information pushing method and device, electronic equipment and storage medium
CN111078742A (en) User classification model training method, user classification method and device
Alhafedh et al. Two-stage gene selection in microarray dataset using fuzzy mutual information and binary particle swarm optimization
CN113656699B (en) User feature vector determining method, related equipment and medium
CN113011886B (en) Method and device for determining account type and electronic equipment
CN111597469A (en) Display position determining method and device, electronic equipment and storage medium
KR102223640B1 (en) Cloud-based personalized contents subscription service providing system and method thereof
CN112651790A (en) OCPX self-adaptive learning method and system based on user reach in fast-moving industry
CN112256969A (en) Media material pushing method and device, electronic equipment and storage medium
CN111461188A (en) Target service control method, device, computing equipment and storage medium
CN108804492A (en) The method and device recommended for multimedia object
CN113139085B (en) User interest mining method and system
CN113392309A (en) Data processing method, device, equipment and storage medium
CN113434560A (en) Sorting method, electronic equipment and related product
CN112732874A (en) Training method and device for user identification call-through model

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
CB02 Change of applicant information
CB02 Change of applicant information

Address after: 100041 B-0035, 2 floor, 3 building, 30 Shixing street, Shijingshan District, Beijing.

Applicant after: Douyin Vision Co.,Ltd.

Address before: 100041 B-0035, 2 floor, 3 building, 30 Shixing street, Shijingshan District, Beijing.

Applicant before: Tiktok vision (Beijing) Co.,Ltd.

Address after: 100041 B-0035, 2 floor, 3 building, 30 Shixing street, Shijingshan District, Beijing.

Applicant after: Tiktok vision (Beijing) Co.,Ltd.

Address before: 100041 B-0035, 2 floor, 3 building, 30 Shixing street, Shijingshan District, Beijing.

Applicant before: BEIJING BYTEDANCE NETWORK TECHNOLOGY Co.,Ltd.

GR01 Patent grant
GR01 Patent grant