CN113177160B - Push text generation method and device, electronic equipment and storage medium - Google Patents
Push text generation method and device, electronic equipment and storage medium Download PDFInfo
- Publication number
- CN113177160B CN113177160B CN202110571716.7A CN202110571716A CN113177160B CN 113177160 B CN113177160 B CN 113177160B CN 202110571716 A CN202110571716 A CN 202110571716A CN 113177160 B CN113177160 B CN 113177160B
- Authority
- CN
- China
- Prior art keywords
- information
- pushed
- target user
- keyword information
- keyword
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 46
- 230000000694 effects Effects 0.000 claims abstract description 33
- 238000012216 screening Methods 0.000 claims description 22
- 230000006870 function Effects 0.000 claims description 18
- 238000004891 communication Methods 0.000 claims description 13
- 238000004590 computer program Methods 0.000 claims description 11
- 238000013507 mapping Methods 0.000 claims description 7
- 238000004422 calculation algorithm Methods 0.000 claims description 6
- 238000012549 training Methods 0.000 claims description 5
- 238000012163 sequencing technique Methods 0.000 claims description 3
- 238000004364 calculation method Methods 0.000 description 14
- 238000010586 diagram Methods 0.000 description 10
- 230000008569 process Effects 0.000 description 6
- 230000009471 action Effects 0.000 description 3
- 230000006399 behavior Effects 0.000 description 3
- 238000001914 filtration Methods 0.000 description 2
- 230000002093 peripheral effect Effects 0.000 description 2
- 238000012545 processing Methods 0.000 description 2
- 239000007787 solid Substances 0.000 description 2
- 239000013598 vector Substances 0.000 description 2
- 230000002354 daily effect Effects 0.000 description 1
- 238000013500 data storage Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000000802 evaporation-induced self-assembly Methods 0.000 description 1
- 230000003203 everyday effect Effects 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 230000010354 integration Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 239000013307 optical fiber Substances 0.000 description 1
- 239000004065 semiconductor Substances 0.000 description 1
- 238000007619 statistical method Methods 0.000 description 1
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
- G06F40/00—Handling natural language data
- G06F40/10—Text processing
- G06F40/166—Editing, e.g. inserting or deleting
- G06F40/186—Templates
Abstract
The method, the device, the electronic equipment and the storage medium for generating the push text are applied to the technical field of information, and the information to be pushed of a target user is obtained; determining keyword information of information to be pushed and keyword information of interest preference of a target user; matching the keyword information of the information to be pushed with the keyword information of each interest preference of the target user to obtain the keyword information matched with the keyword information of the information to be pushed in the keyword information of each interest preference of the target user; and generating a document to be pushed according to the matched keyword information, and pushing the document to be pushed to the target user. The method and the device can generate corresponding text according to the keyword information of interest preference of the user aiming at different users and send the text to the user, solve the technical problem of poor user experience effect caused by single text form in the prior art, and improve the user experience effect.
Description
Technical Field
The present invention relates to the field of information technologies, and in particular, to a method and apparatus for generating a push document, an electronic device, and a storage medium.
Background
With the popularization of mobile intelligent terminal devices and the rapid development of mobile applications, users often browse news, video and other information through the intelligent terminals. Meanwhile, in order to improve the experience effect of the user, each application always pushes hot spot information to the user in a mode of generating and sending a pushing text, so that the user can browse conveniently.
However, at present, when information is pushed, an application program always pushes the same popular video content to all users, the formats of the pushed title and the file are relatively uniform, and after the users receive the push of the type for a long time, the freshness of the users to the push is slowly reduced, so that the user experience effect is poor.
Disclosure of Invention
The embodiment of the invention aims to provide a method, a device, electronic equipment and a storage medium for generating a push text, so as to achieve the aim of improving the experience effect of receiving the push text by a user. The specific technical scheme is as follows:
In a first aspect of the present invention, there is first provided a push document generation method, including:
obtaining information to be pushed of a target user;
determining keyword information of information to be pushed and keyword information of interest preference of a target user;
matching the keyword information of the information to be pushed with the keyword information of each interest preference of the target user to obtain the keyword information matched with the keyword information of the information to be pushed in the keyword information of each interest preference of the target user;
And generating a document to be pushed according to the matched keyword information, and pushing the document to be pushed to the target user.
Optionally, generating the document to be pushed according to the matched keyword information, and pushing the document to be pushed to the target user, including:
Searching a corresponding preset document generation template through the keyword information of interest preference according to the mapping relation between the keyword information of the preset interest preference and the document generation template;
and generating a to-be-pushed document according to the matched keyword information through a preset document generation template, and pushing the to-be-pushed document to a target user.
Optionally, generating a document to be pushed according to the matched keyword information, and pushing the document to be pushed to the target user, and further including:
Generating a document to be pushed according to the matched keyword information;
Acquiring and establishing an active time probability function corresponding to a target user according to a history record of the target user;
According to the activity time probability function, calculating to obtain a time period with the maximum activity probability of the target user as an activity time period of the target user;
and pushing the to-be-pushed document to the target user in the active time period.
Optionally, matching the keyword information of the information to be pushed with the keyword information of each interest preference of the target user to obtain the keyword information matched with the keyword information of the information to be pushed in the keyword information of each interest preference of the target user, including:
Matching the keyword information of the information to be pushed with the keyword information of each interest preference of the target user to obtain the keyword information which is the same as the keyword information of the information to be pushed in the keyword information of each interest preference of the target user, wherein the keyword information is used as the matched keyword information.
Optionally, obtaining information to be pushed of the target user includes:
Calculating to obtain keyword information of interest preference of the target user according to the history record of the target user through a multidimensional user portrait interest model;
Acquiring keyword information of each information in an information base to be pushed;
And matching the keyword information of interest preference of the target user with the keyword information of information in the information base to be pushed, and selecting the information in the information base to be pushed with the highest matching degree as the information to be pushed.
Optionally, obtaining keyword information of each information in the information base to be pushed includes
Acquiring information screened according to a preset screening rule and/or information uploaded by a target user;
Acquiring and sequencing according to the information screened according to a preset screening rule and/or the update time of the information uploaded by the target user;
and acquiring the keyword information of the first N pieces of information with latest update time in the information screened according to the preset screening rule and/or the information uploaded by the target user.
Optionally, the keyword information of each interest preference of the target user is obtained through calculation according to the history record of the target user by using the multidimensional user portrait interest model, and the method comprises the following steps:
acquiring a history browsing record and/or a history watching record of a target user;
and inputting the historical browsing record and/or the historical viewing record of the target user into a pre-trained multidimensional user image interest model, and calculating to obtain keyword information of each interest preference of the target user.
Optionally, matching the keyword information of each interest preference of the target user with the keyword information of each information in the information base to be pushed, and selecting the information in the information base to be pushed with the highest matching degree as the information to be pushed, including:
Inputting keyword information of interest preferences of a target user and keyword information of information in an information base to be pushed into an algorithm prediction model obtained by training in advance to obtain matching degree scores of the keyword information of interest preferences and the keyword information of the information in the information base to be pushed;
and selecting the information in the information base to be pushed with the highest matching degree as the information to be pushed.
In a second aspect of the present invention, there is also provided a pushing document generating device, the device including:
The information to be pushed acquisition module is used for acquiring information to be pushed of a target user;
The keyword information determining module is used for determining keyword information of information to be pushed and keyword information of interest preference of a target user;
The keyword information matching module is used for matching the keyword information of the information to be pushed with the keyword information of each interest preference of the target user to obtain the keyword information matched with the keyword information of the information to be pushed in the keyword information of each interest preference of the target user;
And the to-be-pushed document generation module is used for generating to-be-pushed documents according to the matched keyword information and pushing the to-be-pushed documents to the target user.
Optionally, the to-be-pushed document generating module includes:
the template searching sub-module is used for searching a corresponding preset document generating template through the keyword information of interest preference according to the mapping relation between the keyword information of the interest preference and the document generating template;
And the document pushing sub-module is used for generating a document to be pushed according to the matched keyword information through a preset document generation template and pushing the document to be pushed to a target user.
Optionally, the to-be-pushed document generating module further includes:
The document generation sub-module is used for generating a document to be pushed according to the matched keyword information;
the probability function creation submodule is used for acquiring and establishing an active time probability function corresponding to the target user according to the history record of the target user;
The active time period calculation submodule is used for calculating and obtaining a time period with the maximum active probability of the target user as an active time period of the target user according to the active time probability function;
and the document pushing sub-module is used for pushing the document to be pushed to the target user in the active time period.
Optionally, the keyword information matching module is specifically configured to: matching the keyword information of the information to be pushed with the keyword information of each interest preference of the target user to obtain the keyword information which is the same as the keyword information of the information to be pushed in the keyword information of each interest preference of the target user, wherein the keyword information is used as the matched keyword information.
Optionally, the information obtaining module to be pushed includes:
The interest preference calculation sub-module is used for calculating keyword information of each interest preference of the target user according to the history record of the target user through the multidimensional user portrait interest model;
The keyword information acquisition sub-module is used for acquiring keyword information of each information in the information base to be pushed;
The information to be pushed is selected from the sub-module, which is used for matching the keyword information of interest preference of the target user with the keyword information of information in the information base to be pushed, and selecting the information in the information base to be pushed with the highest matching degree as the information to be pushed.
Optionally, the keyword information acquisition sub-module comprises
The screening and uploading information acquisition sub-module is used for acquiring information screened according to preset screening rules and/or information uploaded by a target user;
The screening and uploading information ordering sub-module is used for acquiring and ordering according to the information screened according to the preset screening rule and/or the update time of the information uploaded by the target user;
And the screening and uploading information selecting sub-module is used for acquiring the keyword information of the last N pieces of information in the information screened according to the preset screening rule and/or the information uploaded by the target user.
Optionally, the interest preference calculation sub-module includes:
The browsing and watching record acquisition sub-module is used for acquiring a historical browsing record and/or a historical watching record of the target user;
the interest preference keyword acquisition sub-module is used for inputting the historical browsing record and/or the historical viewing record of the target user into a pre-trained multidimensional user image interest model, and calculating to obtain keyword information of each interest preference of the target user.
Optionally, the information to be pushed selecting sub-module includes:
The matching degree score calculation sub-module is used for inputting the keyword information of each interest preference of the target user and the keyword information of each information in the information base to be pushed into an algorithm prediction model which is obtained by training in advance to obtain the matching degree score of the keyword information of each interest preference and the keyword information of each information in the information base to be pushed;
The highest matching degree selecting sub-module is used for selecting information in the information base to be pushed with the highest matching degree as information to be pushed.
In yet another aspect of the present invention, there is also provided a computer readable storage medium having a computer program stored therein, which when executed by a processor implements any of the above-described push document generation methods.
In yet another 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 any of the push document generation methods described above.
The embodiment of the invention provides a pushing document generation method, a pushing document generation device, electronic equipment and a storage medium, wherein information to be pushed of a target user is obtained; determining keyword information of information to be pushed and keyword information of interest preference of a target user; matching the keyword information of the information to be pushed with the keyword information of each interest preference of the target user to obtain the keyword information matched with the keyword information of the information to be pushed in the keyword information of each interest preference of the target user; and generating a document to be pushed according to the matched keyword information, and pushing the document to be pushed to the target user. The method and the device can generate corresponding text according to the keyword information of interest preference of the user aiming at different users and send the text to the user, solve the technical problem of poor user experience effect caused by single text form in the prior art, and improve the user experience effect.
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 schematic flow chart of a first method for generating a push document according to an embodiment of the present application;
FIG. 2a is a diagram of a first example of a push document according to an embodiment of the present application;
FIG. 2b is a diagram of a second example of a push document according to an embodiment of the present application;
FIG. 2c is a diagram of a third example of a push document according to an embodiment of the present application;
fig. 3 is a schematic flow chart of pushing to a user according to an embodiment of the present application;
fig. 4 is a schematic flow chart of obtaining information to be pushed of a target user according to an embodiment of the present application;
FIG. 5 is a diagram illustrating an example of a method for generating a push document according to an embodiment of the present application;
fig. 6 is a schematic structural diagram of a pushing document generating device according to an embodiment of the present application;
Fig. 7 is a schematic structural diagram of an electronic device according to an embodiment of the present application.
Detailed Description
The technical solutions in the embodiments of the present invention will be described below with reference to the accompanying drawings in the embodiments of the present invention.
In order to improve the experience effect of receiving the push text by the user, the embodiment of the application provides a push text generation method.
In one embodiment of the present application, the method includes:
obtaining information to be pushed of a target user;
determining keyword information of information to be pushed and keyword information of interest preference of a target user;
matching the keyword information of the information to be pushed with the keyword information of each interest preference of the target user to obtain the keyword information matched with the keyword information of the information to be pushed in the keyword information of each interest preference of the target user;
And generating a document to be pushed according to the matched keyword information, and pushing the document to be pushed to the target user.
Therefore, by the pushing document generation method, corresponding documents can be generated according to the keyword information of interest preference of the user aiming at different users and sent to the user, so that the technical problem of poor user experience effect caused by single document form in the prior art is solved, and the user experience effect is improved.
The pushing document generating method provided by the embodiment of the application is described in detail by a specific embodiment.
The push text in the embodiment of the application can be generated when the application sends push information to the user, and the user can know the information generated after the interest preference through the push text. For example, when pushing a certain chinese song to a user, the pushing text is: you guess that you like chinese music.
Specifically, referring to fig. 1, fig. 1 is a schematic flow chart of a first method for generating a push document according to an embodiment of the present application, including:
step S11, obtaining information to be pushed of a target user.
The target users may be all or part of the users in the current user group. In the actual use process, the user who installs the current application and starts the information push function can be selected from the current user group to serve as a target user. For example, the current user population of the application is initially filtered, and users for which push notifications are closed and users for which the application is uninstalled are filtered. By selecting the target user from the current user group, the effectiveness of information pushing can be improved, and the invalid pushing amount can be reduced.
The information to be pushed can be information obtained by matching according to interest preferences of the target user. For example, by acquiring keyword information of interest preference of a target user, matching the keyword information of interest preference with the keyword information of each information in the information base to be pushed, and taking the information in the information base to be pushed obtained by matching as the information to be pushed. In the actual use process, the keyword information of interest preference of the target user can be obtained through a multidimensional user image interest model according to the history record of the user. The information in the information base to be pushed can be pre-screened information and/or information uploaded by the target user. For example, the uploaded information can be screened manually to screen out popular information and target information uploaded by target users, and the target users can be predetermined appointed users, such as popular uploading users with more fans, and the like.
The pushing document generating method in the embodiment of the application can be executed by an intelligent terminal, and the intelligent terminal can be a computer or a server.
Step S12, determining keyword information of information to be pushed and keyword information of interest preference of a target user.
The keyword information of the information to be pushed can be determined, and the name of each information in the information base to be pushed and the keyword information of related information such as an author and the like can be obtained. For example, when each information in the information base to be pushed is video information, a title content tag, a video type tag, uploading personnel information, a video embedding (embedded) vector, and the like of the video can be acquired.
The keyword information of interest preference of the target user is determined, and the keyword information of preference of the tag interest of the target user, the keyword information of interest preference of the channel category, the keyword information of interest preference of uploading personnel and the keyword information of single video preference can be obtained through calculation according to the history record of the target user through the multidimensional user portrait interest model.
Step S13, matching the keyword information of the information to be pushed with the keyword information of each interest preference of the target user to obtain the keyword information matched with the keyword information of the information to be pushed in the keyword information of each interest preference of the target user.
And matching the keyword information of the information to be pushed with the keyword information of each interest preference of the target user, and inputting the keyword information of the information to be pushed and the keyword information of each interest preference of the target user into a pre-trained prediction model for prediction to obtain a matching degree score between the keyword information of the information to be pushed and the keyword information of each interest preference of the target user. The prediction model may be various models for performing matching degree calculation, such as FM (Factorization Machine, factorizer) model algorithm, and the like.
And obtaining keyword information matched with the keyword information of the information to be pushed in the keyword information of each interest preference of the target user, and selecting the keyword information with the matching degree score larger than a preset threshold value as matched keyword information according to the matching degree score between the keyword information of the information to be pushed and the keyword information of each interest preference of the target user. Or selecting the keyword information with the largest matching degree score as the matched keyword information according to the matching degree score between the keyword information of the information to be pushed and the keyword information of each interest preference of the target user.
Optionally, in the actual use process, the keyword information of the information to be pushed is matched with the keyword information of each interest preference of the target user, and the keyword information which is the same as the keyword information of the information to be pushed in the keyword information of each interest preference of the target user can be selected as the matched keyword information.
Step S14, generating a document to be pushed according to the matched keyword information, and pushing the document to be pushed to the target user.
Optionally, generating the document to be pushed according to the matched keyword information, and pushing the document to be pushed to the target user, including: searching a corresponding preset document generation template through the keyword information of interest preference according to the mapping relation between the keyword information of the preset interest preference and the document generation template; and generating a to-be-pushed document according to the matched keyword information through a preset document generation template, and pushing the to-be-pushed document to a target user.
The text to be pushed is generated according to the matched keyword information, and the corresponding text can be selected according to the type of the matched keyword information, for example, the type of the keyword information is a name, a song type, an author and the like, and the corresponding text is selected according to the type of the keyword information.
For example, referring to fig. 2a, fig. 2a is a diagram of a first push text example of an embodiment of the present application, if the matched keyword information is a label "× (name)", it may be determined that the target user has interest preference for the label "×", so that a corresponding push text "guess you focus" may be generated for the target user instead of "hot spot".
For another example, referring to fig. 2b, fig. 2b is a diagram of a second example of a push document according to an embodiment of the present application, if the matched keyword information is "chinese music" as the song category, a push document "guessing you to pay attention to the chinese music" may be generated for the target user.
For another example, referring to fig. 2c, fig. 2c is a diagram of a third example of a push document according to an embodiment of the present application, and if the matched keyword information is a certain uploading user, a push document "updated by people interested in you" may be generated for the target user.
Therefore, by the pushing document generation method, corresponding documents can be generated according to the keyword information of interest preference of the user aiming at different users and sent to the user, so that the technical problem of poor user experience effect caused by single document form in the prior art is solved, and the user experience effect is improved.
Optionally, referring to fig. 3, step S14 generates a document to be pushed according to the matched keyword information, and pushes the document to be pushed to the target user, and further includes:
step S141, generating a document to be pushed according to the matched keyword information.
Step S142, an active time probability function corresponding to the target user is obtained and established according to the history record of the target user.
Step S143, according to the activity time probability function, calculating a time period with the maximum activity probability of the target user as an activity time period of the target user.
Step S144, in the active time period, the to-be-pushed document is pushed to the target user.
Specifically, an active time probability function corresponding to the target user can be established, and the active probability of each time period is obtained through a statistical method according to the history record of the target user. For example, by counting, the user browses 100 times in the past week, and the browsing times from 20 to 21 points are 30 times, the activity probability from 20 to 21 points of the user is 30%, and the browsing times of each time period are counted sequentially, so as to obtain the activity probability of each time period.
And according to the activity time probability function, calculating a time period with the maximum activity probability of the target user as an activity time period of the target user, and pushing the to-be-pushed document to the target user in the activity time period. For example, when the activity probability of each time period in one day is counted to obtain that the activity probability of the target user from 20 points to 21 points is maximum, the 20 points to 21 points can be used as the activity time period, and the to-be-pushed document can be pushed to the target user.
It can be seen that the to-be-pushed document is pushed to the target user in the active time period by acquiring and building the active time probability function corresponding to the target user according to the history record of the target user. The probability that the target user accepts the pushed document can be improved, so that the document pushing efficiency is improved.
Optionally, referring to fig. 4, step S11 obtains information to be pushed of the target user, including:
Step S111, through the multidimensional user portrait interest model, keyword information of interest preference of the target user is obtained through calculation according to the history record of the target user.
Optionally, the keyword information of each interest preference of the target user is obtained through calculation according to the history record of the target user by using the multidimensional user portrait interest model, and the method comprises the following steps: acquiring a history browsing record and/or a history watching record of a target user; and inputting the historical browsing record and/or the historical viewing record of the target user into a pre-trained multidimensional user image interest model, and calculating to obtain keyword information of each interest preference of the target user.
The history record may include browsing and viewing records. In particular, the browse and view records may include the time of browsing, information browsed, and the like. For example, the user viewed a television series from 20:28 on 12 months in 2020 to 21:01 on 12 months in 2020.
Optionally, a multidimensional user image interest model is established by utilizing browsing and watching histories of the target user in the application, preference of the target user for tag interests, interest preference of channel categories, interest preference of uploaders, single video preference and the like are described from multiple dimensions, and the preference degree of the target user is represented by scores; meanwhile, according to daily user behaviors, the interest preference scores of the user in each dimension are updated, dynamic update of the user portrait interest model is maintained, time attenuation can be added, and if a certain interest preference score exists for the user, but related historical record behaviors are not performed for a long time, the interest point preference score is reduced.
Step S112, obtaining the keyword information of each information in the information base to be pushed.
Optionally, obtaining keyword information of each information in the information base to be pushed includes: acquiring information screened according to a preset screening rule and/or information uploaded by a target user; acquiring and sequencing according to the information screened according to a preset screening rule and/or the update time of the information uploaded by the target user; and acquiring the keyword information of the first N pieces of information with latest update time in the information screened according to the preset screening rule and/or the information uploaded by the target user.
Step S113, matching the keyword information of interest preference of the target user with the keyword information of information in the information base to be pushed, and selecting the information in the information base to be pushed with the highest matching degree as the information to be pushed.
Optionally, matching the keyword information of each interest preference of the target user with the keyword information of each information in the information base to be pushed, and selecting the information in the information base to be pushed with the highest matching degree as the information to be pushed, including: inputting keyword information of interest preferences of a target user and keyword information of information in an information base to be pushed into an algorithm prediction model obtained by training in advance to obtain matching degree scores of the keyword information of interest preferences and the keyword information of the information in the information base to be pushed; and selecting the information in the information base to be pushed with the highest matching degree as the information to be pushed.
It can be seen that keyword information of interest preference of the target user is obtained through calculation according to the history record of the target user, the keyword information of interest preference of the target user is matched with the keyword information of information in the information base to be pushed, and information in the information base to be pushed with the highest corresponding matching degree is selected as the information to be pushed. The text corresponding to the push video can be generated according to the interest preference information of different users and sent to the users.
Referring to fig. 5, fig. 5 is a diagram illustrating an example of a push document generating method according to an embodiment of the present application, including:
1. Selecting candidate videos, marking the video to be pushed in a video library through manual screening, and simultaneously, automatically marking the video of an uploading user set as a white list with the mark to be pushed. And selecting the latest twenty-thousand videos as candidate push videos every day.
2. And defining candidate groups, selecting a user group which opens the mobile application recently, performing preliminary filtering on the user group, filtering out users with closed pushing notices and users for unloading the application, and reducing invalid pushing amount.
3. And generating video features, namely selecting title content tags, video type tags, uploaders, video embedding vectors and the like of the candidate videos as features of the video side.
4. User portrait interest generation, utilizing browsing and viewing history records of users in an application to establish a multidimensional user portrait interest model, describing interest preferences of the users for tags, interest preferences of channel categories, interest preferences of uploaders, single video preferences and the like from multiple dimensions, using scores to represent the preference degrees of the users, adding time interest attenuation, and maintaining dynamic update of the user portrait interest model.
5. And generating a user push list, matching the multidimensional interests of each user with corresponding video features, inputting a pre-trained ordering model to obtain a matching score of each user on each video to be pushed, and selecting the video with the highest score as the final push video.
6. And generating a user pushing document, wherein after the video to be pushed of each user is obtained in the personalized video pushing channel, the user can inquire which interest preference of the user is matched with and recalled. According to the source of video recall, personalized push documents can be matched for users. For the generation relation between the push text and the video to be pushed by the user, a text type database is maintained, and the database contains the mapping relation between the text template and the source of the push video. After personalized push videos are generated for users, a corresponding push document template is found by inquiring the matching sources of the videos to be pushed and the interests of the users, and personalized push documents are generated for each user.
7. After the user is matched with the video, a personalized active time probability function is established for each user by utilizing the time record of the historical behavior data of the user in the application, and the video is issued to the user side in the active time period with the maximum probability of the user, so that the issuing time is dispersed, the probability of disturbing the user is reduced as much as possible, and the pushing effect is improved.
In a second aspect of the present invention, there is also provided a pushing document generating apparatus, referring to fig. 6, the apparatus includes:
The information to be pushed acquisition module 601 is configured to acquire information to be pushed of a target user;
the keyword information determining module 602 is configured to determine keyword information of information to be pushed and keyword information of interest preference of a target user;
The keyword information matching module 603 is configured to match the keyword information of the information to be pushed with the keyword information of each interest preference of the target user, so as to obtain keyword information matched with the keyword information of the information to be pushed in the keyword information of each interest preference of the target user;
The to-be-pushed document generation module 604 is configured to generate a to-be-pushed document according to the matched keyword information, and push the to-be-pushed document to the target user.
Optionally, the to-be-pushed document generating module 604 includes:
the template searching sub-module is used for searching a corresponding preset document generating template through the keyword information of interest preference according to the mapping relation between the keyword information of the interest preference and the document generating template;
And the document pushing sub-module is used for generating a document to be pushed according to the matched keyword information through a preset document generation template and pushing the document to be pushed to a target user.
Optionally, the to-be-pushed document generating module 604 further includes:
The document generation sub-module is used for generating a document to be pushed according to the matched keyword information;
the probability function creation submodule is used for acquiring and establishing an active time probability function corresponding to the target user according to the history record of the target user;
The active time period calculation submodule is used for calculating and obtaining a time period with the maximum active probability of the target user as an active time period of the target user according to the active time probability function;
and the document pushing sub-module is used for pushing the document to be pushed to the target user in the active time period.
Optionally, the keyword information matching module 603 is specifically configured to: matching the keyword information of the information to be pushed with the keyword information of each interest preference of the target user to obtain the keyword information which is the same as the keyword information of the information to be pushed in the keyword information of each interest preference of the target user, wherein the keyword information is used as the matched keyword information.
Optionally, the information to be pushed obtaining module 601 includes:
The interest preference calculation sub-module is used for calculating keyword information of each interest preference of the target user according to the history record of the target user through the multidimensional user portrait interest model;
The keyword information acquisition sub-module is used for acquiring keyword information of each information in the information base to be pushed;
The information to be pushed is selected from the sub-module, which is used for matching the keyword information of interest preference of the target user with the keyword information of information in the information base to be pushed, and selecting the information in the information base to be pushed with the highest matching degree as the information to be pushed.
Optionally, the keyword information acquisition sub-module comprises
The screening and uploading information acquisition sub-module is used for acquiring information screened according to preset screening rules and/or information uploaded by a target user;
The screening and uploading information ordering sub-module is used for acquiring and ordering according to the information screened according to the preset screening rule and/or the update time of the information uploaded by the target user;
And the screening and uploading information selecting sub-module is used for acquiring the keyword information of the last N pieces of information in the information screened according to the preset screening rule and/or the information uploaded by the target user.
Optionally, the interest preference calculation sub-module includes:
The browsing and watching record acquisition sub-module is used for acquiring a historical browsing record and/or a historical watching record of the target user;
the interest preference keyword acquisition sub-module is used for inputting the historical browsing record and/or the historical viewing record of the target user into a pre-trained multidimensional user image interest model, and calculating to obtain keyword information of each interest preference of the target user.
Optionally, the information to be pushed selecting sub-module includes:
The matching degree score calculation sub-module is used for inputting the keyword information of each interest preference of the target user and the keyword information of each information in the information base to be pushed into an algorithm prediction model which is obtained by training in advance to obtain the matching degree score of the keyword information of each interest preference and the keyword information of each information in the information base to be pushed;
The highest matching degree selecting sub-module is used for selecting information in the information base to be pushed with the highest matching degree as information to be pushed.
Therefore, by means of the pushing document generation device, corresponding documents can be generated according to the keyword information of interest preference of the user aiming at different users and sent to the user, the technical problem that in the prior art, the user experience effect is poor due to the fact that the document form is single is solved, and the user experience effect is improved.
The embodiment of the present invention further provides an electronic device, as shown in fig. 7, including a processor 701, a communication interface 702, a memory 703 and a communication bus 704, where the processor 701, the communication interface 702, and the memory 703 perform communication with each other through the communication bus 704,
A memory 703 for storing a computer program;
the processor 701 is configured to execute the program stored in the memory 703, and implement the following steps:
obtaining information to be pushed of a target user;
determining keyword information of information to be pushed and keyword information of interest preference of a target user;
matching the keyword information of the information to be pushed with the keyword information of each interest preference of the target user to obtain the keyword information matched with the keyword information of the information to be pushed in the keyword information of each interest preference of the target user;
And generating a document to be pushed according to the matched keyword information, and pushing the document to be pushed to the target user.
The communication bus mentioned by the above terminal may be a peripheral component interconnect standard (PERIPHERAL COMPONENT INTERCONNECT, abbreviated as PCI) bus or an extended industry standard architecture (Extended Industry Standard Architecture, abbreviated as EISA) bus, etc. The communication bus may be classified as an address bus, a data bus, a control bus, or the like. For ease of illustration, the figures are shown with only one bold line, but not with only one bus or one type of bus.
The communication interface is used for communication between the terminal and other devices.
The memory may include random access memory (Random Access Memory, RAM) or may include 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 aforementioned processor.
The processor may be a general-purpose processor, including a central processing unit (Central Processing Unit, abbreviated as CPU), a network processor (Network Processor, abbreviated as NP), etc.; but may also be a digital signal Processor (DIGITAL SIGNAL Processor, DSP), application Specific Integrated Circuit (ASIC), field-Programmable gate array (FPGA) or other Programmable logic device, discrete gate or transistor logic device, discrete hardware components.
In yet another embodiment of the present invention, a computer readable storage medium is provided, where a computer program is stored, the computer program implementing the push document generation method according to any one of the above embodiments when executed by a processor.
In yet another embodiment of the present invention, a computer program product comprising instructions, which when run on a computer, causes the computer to perform the push document generation method of any of the above embodiments is also provided.
In the above embodiments, it may be implemented in whole or in part 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, produces a flow or function in accordance with embodiments of the present invention, in whole or in part. The computer may be a general purpose computer, a special purpose computer, a computer network, or other programmable apparatus. The computer instructions may be stored in or transmitted from one computer-readable storage medium to another, for example, by wired (e.g., coaxial cable, optical fiber, digital Subscriber Line (DSL)), or wireless (e.g., infrared, wireless, microwave, etc.). The computer readable storage medium may be any available medium that can be accessed by a computer or a data storage device such as a server, data center, etc. that contains an integration of one or more available media. The usable medium may be a magnetic medium (e.g., floppy disk, hard disk, tape), an optical medium (e.g., DVD), or a semiconductor medium (e.g., solid state disk Solid STATE DISK (SSD)), etc.
It is noted that relational terms such as first and second, and the like are 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. Moreover, 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 one … …" does not exclude the presence of other like elements in a process, method, article, or apparatus that comprises the element.
In this specification, each embodiment is described in a related manner, and identical and similar parts of each embodiment are all referred to each other, and each embodiment mainly describes differences from other embodiments. In particular, for apparatus, electronic devices, storage media, computer program product embodiments, the description is relatively simple, as it is substantially similar to method embodiments, with reference to the description of method embodiments in part.
The foregoing description is only of the preferred embodiments of the present invention and is not intended to limit the scope of the present invention. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention are included in the protection scope of the present invention.
Claims (10)
1. A push document generation method, the method comprising:
obtaining information to be pushed of a target user;
Determining keyword information of the information to be pushed and keyword information of interest preference of the target user;
Matching the keyword information of the information to be pushed with the keyword information of each interest preference of the target user to obtain the keyword information matched with the keyword information of the information to be pushed in the keyword information of each interest preference of the target user;
Generating a document to be pushed according to the matched keyword information, and pushing the document to be pushed to the target user;
The generating a document to be pushed according to the matched keyword information, and pushing the document to be pushed to the target user includes:
Searching a corresponding preset document generation template through the keyword information of interest preference according to the mapping relation between the keyword information of the interest preference and the document generation template;
and generating a to-be-pushed document according to the matched keyword information through the preset document generation template, and pushing the to-be-pushed document to the target user.
2. The method of claim 1, wherein the generating a document to be pushed according to the matched keyword information and pushing the document to be pushed to the target user further comprises:
generating a document to be pushed according to the matched keyword information;
acquiring and establishing an active time probability function corresponding to the target user according to the history record of the target user;
According to the activity time probability function, calculating a time period with the maximum activity probability of the target user as an activity time period of the target user;
and pushing the to-be-pushed document to the target user in the active time period.
3. The method according to claim 1, wherein the matching the keyword information of the information to be pushed with the keyword information of each interest preference of the target user to obtain the keyword information matched with the keyword information of the information to be pushed in the keyword information of each interest preference of the target user, includes:
and matching the keyword information of the information to be pushed with the keyword information of each interest preference of the target user to obtain the keyword information which is the same as the keyword information of the information to be pushed in the keyword information of each interest preference of the target user, wherein the keyword information is used as the matched keyword information.
4. The method of claim 1, wherein the obtaining the information to be pushed of the target user includes:
calculating keyword information of interest preference of the target user according to the history record of the target user through a multidimensional user portrait interest model;
Acquiring keyword information of each information in an information base to be pushed;
And matching the keyword information of each interest preference of the target user with the keyword information of each information in the information base to be pushed, and selecting the information in the information base to be pushed with the highest corresponding matching degree as the information to be pushed.
5. The method of claim 4, wherein the obtaining keyword information of each information in the information base to be pushed comprises
Acquiring information screened according to a preset screening rule and/or information uploaded by a target user;
acquiring and sequencing according to the information screened according to the preset screening rule and/or the update time of the information uploaded by the target user;
And acquiring the keyword information of the first N pieces of information with latest update time in the information screened according to the preset screening rule and/or the information uploaded by the target user.
6. The method of claim 4, wherein the calculating, by the multidimensional user portrayal interest model, keyword information of interest preferences of the target user based on the history of the target user comprises:
acquiring a history browsing record and/or a history watching record of the target user;
and inputting the historical browsing record and/or the historical viewing record of the target user into a pre-trained multidimensional user image interest model, and calculating to obtain keyword information of each interest preference of the target user.
7. The method of claim 4, wherein the matching the keyword information of each interest preference of the target user with the keyword information of each information in the information base to be pushed, and selecting the information in the information base to be pushed with the highest matching degree as the information to be pushed, includes:
Inputting the keyword information of each interest preference of the target user and the keyword information of each information in the information base to be pushed into an algorithm prediction model which is obtained by training in advance to obtain a matching degree score of the keyword information of each interest preference and the keyword information of each information in the information base to be pushed;
and selecting the information in the information base to be pushed with the highest matching degree as the information to be pushed.
8. A push document generation device, the device comprising:
The information to be pushed acquisition module is used for acquiring information to be pushed of a target user;
The keyword information determining module is used for determining the keyword information of the information to be pushed and the keyword information of interest preference of the target user;
The keyword information matching module is used for matching the keyword information of the information to be pushed with the keyword information of each interest preference of the target user to obtain the keyword information matched with the keyword information of the information to be pushed in the keyword information of each interest preference of the target user;
the to-be-pushed document generation module is used for generating a to-be-pushed document according to the matched keyword information and pushing the to-be-pushed document to the target user;
The document generation module to be pushed comprises:
the template searching sub-module is used for searching a corresponding preset document generating template through the keyword information of interest preference according to the mapping relation between the keyword information of the interest preference and the document generating template;
and the document pushing sub-module is used for generating a document to be pushed according to the matched keyword information through the preset document generation template and pushing the document to be pushed to the target user.
9. The electronic equipment is characterized by comprising a processor, a communication interface, a memory and a communication bus, wherein the processor, the communication interface and the memory are communicated with each other through the communication bus;
A memory for storing a computer program;
a processor for carrying out the method steps of any one of claims 1-7 when executing a program stored on a memory.
10. A computer-readable storage medium, characterized in that the computer-readable storage medium has stored therein a computer program which, when executed by a processor, implements the method steps of any of claims 1-7.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110571716.7A CN113177160B (en) | 2021-05-25 | 2021-05-25 | Push text generation method and device, electronic equipment and storage medium |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110571716.7A CN113177160B (en) | 2021-05-25 | 2021-05-25 | Push text generation method and device, electronic equipment and storage medium |
Publications (2)
Publication Number | Publication Date |
---|---|
CN113177160A CN113177160A (en) | 2021-07-27 |
CN113177160B true CN113177160B (en) | 2024-04-23 |
Family
ID=76928210
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202110571716.7A Active CN113177160B (en) | 2021-05-25 | 2021-05-25 | Push text generation method and device, electronic equipment and storage medium |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN113177160B (en) |
Families Citing this family (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN113643070A (en) * | 2021-08-20 | 2021-11-12 | 林秀珍 | Intelligent information pushing method and system based on big data |
CN113744066B (en) * | 2021-09-01 | 2024-04-02 | 杭州隆埠科技有限公司 | Information pushing method and device |
CN114722188A (en) * | 2022-06-09 | 2022-07-08 | 北京鑫宇创世科技有限公司 | Advertisement generation method, device and equipment based on operation data and storage medium |
Citations (19)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
EP1353485A2 (en) * | 2002-03-28 | 2003-10-15 | Nokia Corporation | User interface for pushing information in a short range radio mobile terminal |
CN106227786A (en) * | 2016-07-19 | 2016-12-14 | 百度在线网络技术(北京)有限公司 | Method and apparatus for pushed information |
CN106407425A (en) * | 2016-09-27 | 2017-02-15 | 北京百度网讯科技有限公司 | A method and a device for information push based on artificial intelligence |
CN107291755A (en) * | 2016-04-01 | 2017-10-24 | 中国移动通信有限公司研究院 | A kind of terminal method for pushing and device |
CN107943775A (en) * | 2017-11-24 | 2018-04-20 | 泰康保险集团股份有限公司 | Method, apparatus, storage medium and the electronic equipment of message push |
CN108306814A (en) * | 2017-08-11 | 2018-07-20 | 腾讯科技(深圳)有限公司 | Information-pushing method, device, terminal based on instant messaging and storage medium |
CN108563753A (en) * | 2018-04-16 | 2018-09-21 | 深圳壹账通智能科技有限公司 | Message pushes generation method, device and the computer readable storage medium of official documents and correspondence |
CN108900612A (en) * | 2018-06-29 | 2018-11-27 | 百度在线网络技术(北京)有限公司 | Method and apparatus for pushed information |
KR101958936B1 (en) * | 2018-01-23 | 2019-03-19 | 주식회사 이노피아테크 | Method and system for constructing content of interest in a tv channel |
CN110209941A (en) * | 2019-06-03 | 2019-09-06 | 北京卡路里信息技术有限公司 | Method, method for pushing, device, medium and the server of maintenance push content pool |
CN110365497A (en) * | 2019-05-29 | 2019-10-22 | 平安科技(深圳)有限公司 | Information push method, device, electronic equipment and storage medium |
KR102098537B1 (en) * | 2019-08-05 | 2020-04-07 | 주식회사 네오인터랙티브 | Advertisement service system capable of providing advertisement material template for auto multilink based on analyzing bigdata and advertisement method thereof |
CN111198986A (en) * | 2019-12-17 | 2020-05-26 | 中国平安人寿保险股份有限公司 | Information sending method and device, electronic equipment and storage medium |
CN111768234A (en) * | 2020-06-28 | 2020-10-13 | 百度在线网络技术(北京)有限公司 | Method and device for generating recommended case for user, electronic device and medium |
CN111797315A (en) * | 2020-06-29 | 2020-10-20 | 深圳市锦囊互动技术有限公司 | Message pushing method, device, server and medium |
CN111797353A (en) * | 2020-07-03 | 2020-10-20 | 北京字节跳动网络技术有限公司 | Information pushing method and device and electronic equipment |
CN112182197A (en) * | 2020-11-09 | 2021-01-05 | 北京明略软件系统有限公司 | Method, device and equipment for recommending dialect and computer readable medium |
CN112182378A (en) * | 2020-09-28 | 2021-01-05 | 厦门美柚股份有限公司 | Message pushing method, device, terminal and medium |
CN112202663A (en) * | 2020-09-28 | 2021-01-08 | 厦门美柚股份有限公司 | Message pushing method, device, terminal and medium |
Family Cites Families (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP5395461B2 (en) * | 2009-02-27 | 2014-01-22 | 株式会社東芝 | Information recommendation device, information recommendation method, and information recommendation program |
US11093885B2 (en) * | 2019-01-22 | 2021-08-17 | International Business Machines Corporation | Operations augmented enterprise collaborative recommender engine |
-
2021
- 2021-05-25 CN CN202110571716.7A patent/CN113177160B/en active Active
Patent Citations (19)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
EP1353485A2 (en) * | 2002-03-28 | 2003-10-15 | Nokia Corporation | User interface for pushing information in a short range radio mobile terminal |
CN107291755A (en) * | 2016-04-01 | 2017-10-24 | 中国移动通信有限公司研究院 | A kind of terminal method for pushing and device |
CN106227786A (en) * | 2016-07-19 | 2016-12-14 | 百度在线网络技术(北京)有限公司 | Method and apparatus for pushed information |
CN106407425A (en) * | 2016-09-27 | 2017-02-15 | 北京百度网讯科技有限公司 | A method and a device for information push based on artificial intelligence |
CN108306814A (en) * | 2017-08-11 | 2018-07-20 | 腾讯科技(深圳)有限公司 | Information-pushing method, device, terminal based on instant messaging and storage medium |
CN107943775A (en) * | 2017-11-24 | 2018-04-20 | 泰康保险集团股份有限公司 | Method, apparatus, storage medium and the electronic equipment of message push |
KR101958936B1 (en) * | 2018-01-23 | 2019-03-19 | 주식회사 이노피아테크 | Method and system for constructing content of interest in a tv channel |
CN108563753A (en) * | 2018-04-16 | 2018-09-21 | 深圳壹账通智能科技有限公司 | Message pushes generation method, device and the computer readable storage medium of official documents and correspondence |
CN108900612A (en) * | 2018-06-29 | 2018-11-27 | 百度在线网络技术(北京)有限公司 | Method and apparatus for pushed information |
CN110365497A (en) * | 2019-05-29 | 2019-10-22 | 平安科技(深圳)有限公司 | Information push method, device, electronic equipment and storage medium |
CN110209941A (en) * | 2019-06-03 | 2019-09-06 | 北京卡路里信息技术有限公司 | Method, method for pushing, device, medium and the server of maintenance push content pool |
KR102098537B1 (en) * | 2019-08-05 | 2020-04-07 | 주식회사 네오인터랙티브 | Advertisement service system capable of providing advertisement material template for auto multilink based on analyzing bigdata and advertisement method thereof |
CN111198986A (en) * | 2019-12-17 | 2020-05-26 | 中国平安人寿保险股份有限公司 | Information sending method and device, electronic equipment and storage medium |
CN111768234A (en) * | 2020-06-28 | 2020-10-13 | 百度在线网络技术(北京)有限公司 | Method and device for generating recommended case for user, electronic device and medium |
CN111797315A (en) * | 2020-06-29 | 2020-10-20 | 深圳市锦囊互动技术有限公司 | Message pushing method, device, server and medium |
CN111797353A (en) * | 2020-07-03 | 2020-10-20 | 北京字节跳动网络技术有限公司 | Information pushing method and device and electronic equipment |
CN112182378A (en) * | 2020-09-28 | 2021-01-05 | 厦门美柚股份有限公司 | Message pushing method, device, terminal and medium |
CN112202663A (en) * | 2020-09-28 | 2021-01-08 | 厦门美柚股份有限公司 | Message pushing method, device, terminal and medium |
CN112182197A (en) * | 2020-11-09 | 2021-01-05 | 北京明略软件系统有限公司 | Method, device and equipment for recommending dialect and computer readable medium |
Non-Patent Citations (2)
Title |
---|
基于Ontology的个性化推送系统;潘红艳, 林鸿飞, 赵晶;计算机工程与应用;20060801(第20期);179-183 * |
用户行为数据分析下的信息推送系统的设计;沈军彩;;现代电子技术;20170901(17);166-169 * |
Also Published As
Publication number | Publication date |
---|---|
CN113177160A (en) | 2021-07-27 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN113177160B (en) | Push text generation method and device, electronic equipment and storage medium | |
CN109086439B (en) | Information recommendation method and device | |
CN106326391B (en) | Multimedia resource recommendation method and device | |
CN111178970B (en) | Advertisement putting method and device, electronic equipment and computer readable storage medium | |
CN110223186B (en) | User similarity determining method and information recommending method | |
WO2007071143A1 (en) | Method and apparatus for issuing network information | |
US20110161091A1 (en) | Systems and Methods for Connecting Entities Through Content | |
WO2008048993A2 (en) | Method for estimating real life relationships and popularities among people based on personal visual data | |
CN112330398B (en) | Object processing method and device, electronic equipment and storage medium | |
CN110728370B (en) | Training sample generation method and device, server and storage medium | |
US20180060788A1 (en) | System and method for attribute matching | |
CN108470057B (en) | Generating and pushing method, device, terminal, server and medium of integrated information | |
CN110971973A (en) | Video pushing method and device and electronic equipment | |
CN111753201A (en) | Information pushing method, device, terminal and medium | |
CN108596647B (en) | Advertisement putting method and device and electronic equipment | |
CN111324733A (en) | Content recommendation method, device, equipment and storage medium | |
CN110674330B (en) | Expression management method and device, electronic equipment and storage medium | |
CN111353052B (en) | Multimedia object recommendation method and device, electronic equipment and storage medium | |
CN113536138A (en) | Network resource recommendation method and device, electronic equipment and readable storage medium | |
US20170255632A1 (en) | System and method for identifying trending content based on context | |
TWM617933U (en) | News and public opinion analysis system | |
CN111080349B (en) | Method, device, server and medium for identifying multiple devices of same user | |
CN112115300A (en) | Text processing method and device, electronic equipment and readable storage medium | |
CN116226523B (en) | Knowledge training method and system based on big data | |
CN109962945B (en) | Information acquisition method |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant |