CN113010790A - Content recommendation method, device, server and storage medium - Google Patents
Content recommendation method, device, server and storage medium Download PDFInfo
- Publication number
- CN113010790A CN113010790A CN202110340317.XA CN202110340317A CN113010790A CN 113010790 A CN113010790 A CN 113010790A CN 202110340317 A CN202110340317 A CN 202110340317A CN 113010790 A CN113010790 A CN 113010790A
- Authority
- CN
- China
- Prior art keywords
- recall
- content
- pool
- recommendation
- historical
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 54
- 230000004044 response Effects 0.000 claims abstract description 10
- 238000012216 screening Methods 0.000 claims description 18
- 238000001914 filtration Methods 0.000 claims description 6
- 238000012163 sequencing technique Methods 0.000 claims description 5
- 230000006399 behavior Effects 0.000 description 10
- 238000010586 diagram Methods 0.000 description 9
- 238000012545 processing Methods 0.000 description 7
- 230000008569 process Effects 0.000 description 5
- 238000004590 computer program Methods 0.000 description 4
- 239000013589 supplement Substances 0.000 description 4
- 239000011435 rock Substances 0.000 description 3
- 230000003287 optical effect Effects 0.000 description 2
- 230000001502 supplementing effect Effects 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 239000007787 solid Substances 0.000 description 1
- 238000012546 transfer Methods 0.000 description 1
- 230000001960 triggered effect Effects 0.000 description 1
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
- G06F16/60—Information retrieval; Database structures therefor; File system structures therefor of audio data
- G06F16/63—Querying
- G06F16/635—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
- G06F16/90—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
Abstract
The application discloses a content recommendation method, a content recommendation device, a server and a storage medium, and relates to the technical field of computers. The method comprises the following steps: receiving a content recommendation request sent by a terminal, wherein the content recommendation request comprises user information; acquiring first recall content from a first recall pool based on user information, wherein the first recall pool contains full recall content; in response to the recall amount of the first recall content being smaller than the recall amount threshold, acquiring second recall content from a second recall pool, wherein the recall content in the second recall pool is determined to be obtained based on the historical recall record of the recall content in the first recall pool; and pushing the target recall content to the terminal based on the first recall content and the second recall content. By adopting the scheme provided by the embodiment of the application, the content recall quality can be improved and the probability of using the recalled content by a subsequent user is improved on the premise of ensuring that the sufficient content is recalled.
Description
Technical Field
The embodiment of the application relates to the technical field of computers, in particular to a content recommendation method, a content recommendation device, a server and a storage medium.
Background
The content recommendation refers to selecting content from a recommended content library and recommending the content to a user, and generally, in the process of selecting recommended content, the user characteristics and historical behavior data need to be combined, and content which is possibly interested by the user is recalled from the recommended content library based on the user characteristics and the historical behavior data.
In the prior art, when a recommendation system recommends content to a new user or a cold-start user, summary statistics is generally performed according to real-time online or offline received user behaviors, and the recommended content is recalled from a recommendation library, but the cold-start user often has insufficient user behavior data, and the recalled recommended content is insufficient.
Disclosure of Invention
The embodiment of the application provides a content recommendation method, a content recommendation device, a server and a storage medium. The technical scheme is as follows:
in one aspect, an embodiment of the present application provides a content recommendation method, where the method includes:
receiving a content recommendation request sent by a terminal, wherein the content recommendation request comprises user information;
acquiring first recall content from a first recall pool based on the user information, wherein the first recall pool comprises full recall content;
in response to the recall amount of the first recall content being less than a recall amount threshold, obtaining second recall content from a second recall pool, the recall content in the second recall pool determined to be obtained based on historical recall records of recall content in the first recall pool;
and pushing target recall content to the terminal based on the first recall content and the second recall content.
In another aspect, an embodiment of the present application provides a content recommendation apparatus, where the apparatus includes:
the system comprises a receiving module, a sending module and a receiving module, wherein the receiving module is used for receiving a content recommendation request sent by a terminal, and the content recommendation request comprises user information;
the first recall module is used for acquiring first recall content from a first recall pool based on the user information, wherein the first recall pool comprises full recall content;
a second recall module, configured to, in response to a recall amount of the first recall content being less than a recall amount threshold, obtain second recall content from a second recall pool, where the recall content in the second recall pool is determined to be obtained based on a historical recall record of recall content in the first recall pool;
and the pushing module is used for pushing the target recall content to the terminal based on the first recall content and the second recall content.
In another aspect, an embodiment of the present application provides a server, where the live broadcast server includes a processor and a memory, where the memory stores at least one program, and the at least one program is loaded and executed by the processor to implement the content recommendation method in the foregoing aspect.
In another aspect, an embodiment of the present application provides a computer-readable storage medium, which stores at least one instruction for execution by a processor to implement the content recommendation method according to the above aspect.
In another aspect, embodiments of the present application provide a computer program product or a computer program, which includes computer instructions stored in a computer-readable storage medium. The processor of the computer device reads the computer instructions from the computer-readable storage medium, and the processor executes the computer instructions to cause the computer device to execute the content recommendation method provided by the above aspect.
The beneficial effects brought by the technical scheme provided by the embodiment of the application at least comprise:
according to the method provided by the embodiment of the application, the server acquires the user information contained in the content recommendation request according to the acquired content recommendation request sent by the terminal, and further acquires the first recall content from the first recall pool according to the user information; in addition, considering that the cold start stage is only based on the user information, the sufficient content is not recalled from the first recall pool, the first recall content and the second recall content are pushed to the terminal by acquiring the second recall content from the second recall pool, and the sufficient recommended content is pushed to the user; in addition, the second recall content is determined and obtained based on the historical recall record of the recall content in the first recall pool, so by adopting the scheme provided by the embodiment of the application, the content recall quality can be improved and the probability of using the recall content by a subsequent user can be improved on the premise of ensuring that enough content is recalled.
Drawings
FIG. 1 is a diagram illustrating an environment for implementing a content recommendation method provided by an exemplary embodiment of the present application;
FIG. 2 is a flow chart of a method of content recommendation provided by an exemplary embodiment of the present application;
FIG. 3 illustrates a system architecture diagram of a recommendation system provided by an exemplary embodiment of the present application;
FIG. 4 is a flow chart of a method of content recommendation provided by another exemplary embodiment of the present application;
FIG. 5 is a flow chart of a method of content recommendation provided by another exemplary embodiment of the present application;
FIG. 6 is a flow chart of a second recall pool update process provided by an exemplary embodiment of the present application;
FIG. 7 is a diagram illustrating an implementation of a content recommendation method according to an exemplary embodiment of the present application;
fig. 8 is a block diagram of a content recommendation apparatus according to an exemplary embodiment of the present application;
fig. 9 is a block diagram of a server according to an embodiment of the present application.
Detailed Description
To make the objects, technical solutions and advantages of the present application more clear, embodiments of the present application will be described in further detail below with reference to the accompanying drawings.
In the related art, when the recommended content recalled by the cold-start user is insufficient, the recommendation platform mostly adopts a mode of supplementing the content with a higher current click rate or manually configuring the content through operation, however, the popularity of the recommended content cannot be guaranteed in a mode of determining the search volume or the click rate of the content, the real-time performance is poor in the mode of manually configuring through operation, and the process is complicated.
In the scheme provided by the embodiment of the application, the recommendation system selects content to be pushed to the user according to the obtained user information, and online real-time summary statistics is carried out on the content while pushing the content, so that the popularity data of recently recommended content, namely the latest popular recommended data, is obtained. The method mainly solves the problems that the user cannot obtain enough recommended content from the recommendation system and the recommendation of the cold start user, ensures that the content with higher recent recall degree is pushed to the user, improves the content recall quality and improves the probability of using the recalled content by the subsequent user.
Referring to fig. 1, a schematic diagram of an implementation environment provided by an embodiment of the present application is shown. The implementation environment may include: a terminal 110 and a server 120.
The terminal 110 has an application program with a push function installed therein, and the terminal may be an electronic device such as a mobile phone, a desktop computer, a tablet computer, a multimedia playing device, and a laptop portable computer. The application may be a music-like application, a social-like application, a shopping-like application, a news-like application, a video-like application, a gourmet-like application, and the like. In this embodiment, the user logs in a corresponding user account by using the terminal 110, and obtains a corresponding push content by sending a recommendation request to the server 120.
The terminal 110 is connected to the server 120 through a wireless network or a wired network.
The server 120 includes at least one of a server, a plurality of servers, a cloud computing platform, and a virtualization center. The server 120 provides background services for applications in the terminal 110. For example, server 120 may be a backend server for an application as described above. In this embodiment of the application, the server 120 runs with a recommendation system 121, the recommendation system 121 is configured to recall, from among a large amount of contents, contents that need to be recommended according to a recommendation request of a user, and the server 120 may receive the recommendation request sent from the terminal 110, obtain, according to a push request, recommended contents from the recommendation system 121, and send the recommended contents to the terminal 110; optionally, the server 120 is further configured to store account information corresponding to each account. For music software, the server 120 may store historical listening records, historical recommendation records, historical collection records, music purchase records, comments of the user, praise records, and the like of corresponding accounts, the server 120 may recommend songs to the user based on the stored record information, and the server 120 may further store social relationships among the user accounts, determine other user accounts associated with the social relationships according to the social relationships, and further use information of the other accounts as a recommendation basis.
Fig. 2 is a flowchart of a content recommendation method according to an exemplary embodiment of the present application, and this embodiment takes the method as an example for being used in the server in fig. 1 to describe. The method comprises the following steps.
When a user starts an application program to obtain recommended content, a content recommendation request is sent to a server through a trigger terminal. The content recommendation request comprises a user account corresponding to the terminal. When the server receives a content push request sent by the terminal, user information contained in the content push request is acquired, wherein the user information can be an account identifier of a user, the server acquires account information corresponding to the account according to the account identifier, and the account information of the user can include at least one of user characteristic information and user behavior data. The user characteristic information may include gender information, age information, occupation information, preference information, and the like of the user; the user behavior data may include clicking operations and browsing contents of the application program by the user.
Taking a terminal starting a music application program as an example, when a user starts music software and needs to acquire recommended songs, the terminal is triggered to send a recommendation request to a server through clicking a song recommendation control, and the server acquires user information corresponding to the terminal according to the acquired recommendation request sent by the terminal, wherein the user information comprises information such as account information of logging in the music software, a song listening record of the user, a music style preferred by the user and the like.
As shown in fig. 3, a recommendation system 320 runs in the server, the recommendation system 320 is configured to recommend content to the user according to the user information obtained by the server, the recommendation system 320 is composed of a first recall pool 321 and a second recall pool 322, and the first recall pool 321 includes a full amount of recall content.
The recommendation system acquires first recall content from a first recall pool based on the acquired user information, wherein the first recall pool contains full recall content for recommending content to a user, if an application program is music software, a music song library is stored in the corresponding first recall pool, and the first recall content is a song to be recommended based on user information recall; when the application program is news software, a news information base is stored corresponding to the first recall pool, and the first recall content is news to be recommended based on user information recall. The user information may be recall content obtained based on at least one of account information, user characteristic information and user behavior data of the user.
In a possible implementation manner, when a server receives a recommendation request sent by a terminal, a recommendation system classifies users according to acquired account information of the users and age information, gender information, occupation information and the like of the corresponding users, if the user is determined to be '70 back' according to the age information of the users, attributes of recalled songs are positioned as classical songs or opera songs, and if the user age is '00 back', more songs of recommended rock and roll music and record Jockey (DJ) songs are obtained; if songs are recommended according to the gender information of the user, songs sung by a male singer are more recommended when the user is female.
In another possible implementation, the recommendation may be further performed according to the historical song listening records of the user, for example, the historical song listening records of the user in the last month are obtained, the songs in the historical records are classified and ranked, and if the types of the top songs in the song listening records of the user are light music and balladry, a certain proportion or a certain number of the light music and balladry songs are selected from the first recall pool. In addition, the preference of the user can be determined according to the acquired user behavior data of the user to the application program, for example, songs similar to the songs played by the user are acquired from the first recall pool according to the song content of the songs played by the user.
In addition, the relevant songs and other songs of the singer can be obtained from the first recall pool by obtaining the songs collected by the user or the singer concerned, or the first recall content is selected according to the song listening ranking list or the downloading ranking list of the platform.
In a possible scenario, if the user sending the content recommendation request is a cold-start user, that is, a user without historical behavior data, such as a newly registered user, the server cannot acquire user information included in the recommendation request sent by the terminal, or the use time of the registered user is short, the user information data acquired by the server is insufficient to acquire enough first recall content from the first recall pool, that is, the recall amount of the first recall content does not reach the recall amount threshold, and the content cannot be recommended to the user.
As shown in fig. 3, when the recall amount of the first recall content acquired from the first recall pool 321 is less than the set threshold, the second recall content needs to be acquired from the second recall pool 322, so that the recall amount of the recall content reaches the recall amount threshold or exceeds the recall threshold, and it is ensured that the recommended content received by the terminal reaches a certain number, and the recommended content is not too little to affect the viewing of the user. The recalled content in the second recall pool 322 is determined based on the historical recall record recalled from the first recall pool 321, that is, the recalled content stored in the second recall pool 322 is the first recalled content in the historical recall record that meets certain conditions.
Illustratively, the server acquires a certain number of songs to be recommended from the first recall pool for multiple times according to the user information, and stores a part of the songs to be recommended which are recalled for multiple times into the second recall pool, wherein the songs stored in the second recall pool are vividly understood as popular songs recalled from the first recall pool for multiple times.
And step 204, pushing the target recall content to the terminal based on the first recall content and the second recall content.
In one possible implementation, when the sum of the recall amounts of the first recall content and the second recall content reaches a recall amount threshold, it is determined that the recommendation condition is satisfied, the first recall content and the second recall content are determined as target recall content, and the target recall content is pushed to the terminal.
In another possible implementation manner, if the determined recall amount of the first recall content and the second recall content is large, and all the recall content does not need to be pushed to the terminal actually, the first recall content and the second recall content may be further filtered to obtain the filtered target recall content, and then the target recall content is pushed to the terminal.
According to the method provided by the embodiment of the application, the server acquires the user information contained in the content recommendation request according to the acquired content recommendation request sent by the terminal, and further acquires the first recall content from the first recall pool according to the user information; in addition, considering that the cold start stage is only based on the user information, the sufficient content is not recalled from the first recall pool, the first recall content and the second recall content are pushed to the terminal by acquiring the second recall content from the second recall pool, and the sufficient recommended content is pushed to the user; in addition, the second recall content is determined and obtained based on the historical recall record of the recall content in the first recall pool, so by adopting the scheme provided by the embodiment of the application, the content recall quality can be improved and the probability of using the recall content by a subsequent user can be improved on the premise of ensuring that enough content is recalled.
Fig. 4 is a flowchart of a content recommendation method according to another exemplary embodiment of the present application, and this embodiment takes the method as an example for being used in the server shown in fig. 1. The method comprises the following steps:
Step 201 may be referred to in the implementation manner of this step, and this embodiment is not described herein again.
The implementation manner of this step may refer to step 202, which is not described herein.
In one possible implementation, the user information data obtained by the server is insufficient for the cold-start user to recall a sufficient amount of the first recalled content from the first recall pool, that is, the recall amount of the first recalled content is less than the recall amount threshold. At this time, the second recall content is required to be acquired from the second recall pool so that the recall amount of the recall content reaches the recall amount threshold or exceeds the recall threshold. Therefore, as shown in fig. 5, step 403 may further include the following steps.
And determining a second recall quantity of the second recall content required to be recalled from the second recall pool according to the acquired first recall quantity of the first recall content and the set recall quantity threshold value because the first recall quantity of the first recall content acquired by the recommendation system based on the user information is smaller than the recall quantity threshold value. Typically, the second recall amount is inversely related to the first recall amount, i.e., the more the first recall amount of the recalled content, the less the second recall content needs to be obtained. For example, the second recall amount of the second recall content is determined according to a recall amount threshold set by the recommendation system and a difference value of the first recall amount corresponding to the first recall content.
Illustratively, taking the recommendation system to recommend songs to the terminal as an example, the recommendation system recalls 21 songs from the first recall pool based on the user information, but considering the time length of listening to songs per day and the time length of each song by the user, the number of songs recommended to the user set by the recommendation system is set to 30, and since the user information data is not enough to obtain a sufficient number of recommended songs, the recommendation system selects at least 9 songs from the second recall pool as supplementary songs, so that the recommended songs reach or exceed 30 songs, and the push condition is met.
In step 403B, a second recall content of a second recall amount is obtained from the second recall pool.
And determining the recall quantity of the second recall content according to the difference value of the recall quantity threshold and the recall quantity corresponding to the first recall content, and selecting the second recall content of the second recall quantity from the second recall pool.
Because the recall content in the second recall pool is determined and obtained based on the historical recall record of the recall content in the first recall pool, the second recall content selected from the second recall pool is not obtained based on user information acquisition, but the recommendation system is hot recall content determined by the number of times the content is recalled when recommending the content to other users in the past, therefore, when the first recall amount does not exceed the recall threshold, the content with higher recall degree can be provided for the user by supplementing the second recall content, so as to meet the requirement of the user on the hot content.
In a possible implementation manner, the recall amount of the first recall content acquired according to the user information is relatively close to the recall amount threshold, but the numerical value of the second recall amount determined by the difference value between the recall amount threshold and the recall amount corresponding to the first recall content is relatively small, so that the contents pushed to the user in the recommendation manner may all be contents of the same type or the same tag attribute, and as a result, the pushed contents are easily simplified, and the user experience is reduced. Therefore, on the basis of ensuring the second recall amount, the second recall amount is properly increased, and if the second recall amount is set to be integral multiple of the difference value between the recall threshold and the first recall amount, the recall amount of the recall content can be ensured to reach the recall threshold, the probability of pushing the second recall content to the user is increased, the simplification of the pushed content is avoided, and meanwhile, the situation that the recall amount cannot reach the recommendation condition after the recall content is further screened subsequently is avoided.
In step 404, a union of the first recalled content and the second recalled content is determined as a third recalled content.
And after the second recall content is acquired from the second recall pool, comparing the second recall content with the first recall content, if the first recall content and the second recall content have repeated recall content, removing the repeated recall content from the first recall content or the second recall content, ensuring that the sum of the first recall amount and the second recall amount reaches a recall amount threshold value, simultaneously avoiding the repeated recall content, determining the union of the first recall content and the second recall content as a third recall content, and ensuring that the recall amount of the third recall content is not less than the recall amount threshold value.
And after the recommending system determines the third recall content, acquiring a historical recommendation record corresponding to the user information according to the acquired user information, wherein the historical recommendation record recommended to the user within a recent period of time is recorded in the historical recommendation record, for example, the historical recommendation record pushed to the user within a month is stored, filtering the third recall content according to the acquired historical recommendation record, and filtering the recall content in the historical recommendation contained in the third recall content to obtain the target recall content.
Illustratively, taking recommended songs as an example, the server acquires historical song recommendation records of corresponding accounts within one month according to the acquired user account information, and filters pushed songs from the acquired third recalled content according to the historical song recommendation records, for example, the third recalled content includes 40 recalled songs, and filters 5 recommended songs by comparing the third recalled content with the historical push records of the user in the previous month, so that the acquired target recalled content includes 35 target recalled songs.
In addition, the corresponding historical song listening records of the user can be obtained according to the user information, and if songs in the historical song listening records of the user in the last month are detected to be contained in the third recall content, the corresponding songs are filtered. In addition, in order to further improve user experience, copyright information of music in the third recalled content can be acquired, and if the corresponding music software platform does not have the copyright of the corresponding song, the corresponding song is filtered, so that the non-copyright song is prevented from being pushed to the user, and the user experience is reduced; in addition, whether the user is a premium user (VIP) or not can be judged according to the user information, and if the user is not a VIP user, the corresponding VIP song is filtered.
And step 406, updating the historical recommendation record corresponding to the user information based on the target recall content.
And updating the historical recommendation record corresponding to the user information after the determined target recall content is based on the determined target recall content, namely the determined target recall content needs to be recommended to the user, so that the determined target recall content needs to be updated as a historical recommendation record, and repeated recommendation content is avoided when the target recall content is recommended to the user next time.
In another possible implementation, the user may not use the content recommended by the recommendation system, and if the updated recommendation record may be the same as the user missing the recommended content, the pushed and used content may be determined according to the historical usage record of the user by obtaining the historical usage record of the user, and the content may be updated based on the historical usage record.
In addition, the historical song listening records of the user can be used as an updating basis, the historical song listening records of the user in a certain time are obtained before the records are updated, and the corresponding historical song listening records are updated.
After the target recall content is determined, the display sequence of the entry recall content at the terminal needs to be sequenced, and generally, when a user browses the recommended content, the probability that the content with the highest ranking is browsed or clicked by the user is higher than that of the content with the lowest ranking, so that the recommendation priority of the entry recall content needs to be determined.
In one possible implementation, since the first recalled content retrieved from the first recall pool is determined based on user information of the user, the first recalled content is of high probability of being content of interest to the user, while the second recalled content retrieved from the second recall pool is not determined based on user information and is not necessarily of interest to the user. Therefore, the recommendation priority of the target recall content from the first recall pool is set higher than that of the target recall content from the second recall pool, that is, the recommendation sequence of the first recall content is earlier and the recommendation sequence of the second recall content is later.
Illustratively, most of the first recall contents are ballad songs and love songs that are interesting to the user, and most of the second recall contents are rock songs, the ballad songs and love songs are preferentially displayed at the front position of the recommendation list, and the rock songs are arranged at the rear position.
At step 408, a recommendation priority is determined based on the historical recall times for each entry recall content.
In addition to sequencing recommended contents according to user preferences, in order to enrich the display mode of the recommended contents, before target recall contents are recommended to a user, the recall times of the recalled contents of all items in a certain time are obtained, and the recommendation priority is determined according to the recall times, namely the recommendation priority of the contents with high recall times is higher than the recommendation priority of the contents with low recall times.
In addition, the priority can be determined according to the source and the recall times of the recalled contents of each item mark, for example, the recommendation priority of the first recalled content is set to be higher than that of the second recalled content, and the recommendation priority of each item is determined according to the recall times of the second recalled content.
In order to avoid the user from viewing or clicking the content to be singulated, the priority may be determined in accordance with the alphabetical order of the target recalled content, or the priority may be determined randomly.
And 409, sequencing the target recall content based on the recommendation priority, and pushing the sequenced target recall content to the terminal.
And after the recommendation priority of the target recall content is determined, pushing the target recall content to the corresponding terminal.
In the embodiment of the application, when the server is insufficient to obtain enough first recall content from the first recall pool based on the obtained user information data, the second recall content is obtained from the second recall pool for parallel supplement, and the union of the first recall content and the second recall content is determined as the third recall content, so that the recall volume of the recall content reaches a recall volume threshold value, and the bibliographic data is provided under the condition that the recall content is insufficient;
furthermore, the third recall content is screened according to the historical recommendation record and sorted according to the recommendation priority, so that personalized target recall content is recommended to the user, and the phenomenon that the experience of the user is reduced due to the simplification of the recommended content is avoided.
In a possible application scenario, if the user sending the content recommendation request is a cold-start user, the server cannot acquire enough historical behavior data, so that the server cannot acquire enough first recall content from the first recall pool, and therefore needs to acquire second recall content from the second recall pool as the bibliographic data. In order to ensure that the latest popular recommended content with a large number of recalls is obtained from the second recall pool, the recall content in the second recall pool needs to be updated in real time by performing data screening on the first recall content.
Fig. 6 is a flowchart of a second recall pool update process according to another exemplary embodiment of the present application, which is described by taking the method as an example for the server shown in fig. 1. The method comprises the following steps:
And after the recommendation system pushes the target recall content to the user according to the configuration information, acquiring a historical recall record of each recall content in the first recall pool within a certain time, wherein the historical recall record is obtained by counting the target recall content pushed to the user within a set time, and the historical record can comprise the recall time, the recall times, the name of the recall content and the like of the acquired target recall content.
Illustratively, for example, when the recommendation system recalls songs, after the recommendation system recommends the recalled songs to the user, a historical recall record of the songs in the first recall pool is obtained, where the historical recall record may be a historical recall record of the songs recommended to the user in the last week or the last month, and includes names of the songs, the latest recall time of each song, the number of recalls of the songs, and the like.
After the recall content in the second recall pool is determined, the historical recall record is also required to be updated, wherein the historical recall record comprises the historical recall time and the recall times corresponding to the updated target recall content, and the historical recall time and the recall times are the recall time and the recall times corresponding to the latest recall of the corresponding content.
Illustratively, taking a recommended song as an example, after a target recall song is determined and a song is recommended to a user, according to the determined target recall song, a historical recall record of the corresponding song in half an hour is acquired from a first recall pool, wherein the historical recall record includes a recall time and a recall number of the last recall of the song, if the historical recall record of the acquired song a is three minutes before the last recall time and the total recall number is one hundred, after the historical recall record is updated, the last recall time of the corresponding song a is updated to be the current time, and the total recall number is one hundred zero.
Because the first recall pool of the recommendation system contains a large amount of recall content, and the historical recall records of the recommendation system contain more recall records, when the content in the second recall pool needs to be updated, if the historical recall records with longer time period are acquired, the load of the recommendation system for processing data can be increased, and the timeliness and accuracy of the recall content heat degree cannot be satisfied by the recall content in the second recall pool after the update is completed.
The recommendation system counts the recall time and the recall times of all the recall contents in the first recall pool based on the obtained historical recall record, determines the recall contents with the recall times larger than or equal to the minimum recall times in a preset recall time period according to the recall time and the recall times, and determines the recall contents as candidate recall contents.
Illustratively, taking a recommended song as an example, based on the obtained historical recall song records, the latest recall time and recall times of each song in the historical recall song records are obtained, the historical recall songs within the latest 30 minutes and with the recall times of more than 100 are screened, and the screened songs are used as candidate recall songs.
At step 604, the candidate recalled content is ranked based on the number of recalls.
In one possible embodiment, the recommendation system determines more candidate recall contents according to the recall time and the number of recalls, wherein a large amount of contents with earlier time comparison and relatively lower number of recalls may be included, but the contents stored in the second recall pool which need to be updated are recently popular recall contents, so that the candidate recall contents need to be further screened, and the recall contents of the current comparison popular are determined and updated.
When the recommendation system determines the candidate recall contents according to the recall time and the recall times, the candidate recall contents are ranked, and are ranked according to the number of the recall times, namely the recall times of the candidate recall contents are obtained, the candidate recall contents with more candidate recall times are ranked at the front position, the candidate recall contents with less recall times are ranked at the rear position, when the candidate recall contents with the same recall times exist, the most recent recall time of the corresponding candidate recall contents is obtained, the most recently recalled candidate contents are ranked at the front position, and the real-time performance of the recommendation heat degree of the recall contents is ensured.
Illustratively, after the candidate recall songs are recalled by the recommendation system, the number of recalls of each song is determined, the songs are sorted according to the number of recalls, when the number of recalls of at least two songs is the same, if it is detected that the number of recalls of song a and song B in the last thirty minutes is 365, the recall time of the corresponding song is obtained, and if the last recall time of song a is three minutes ago and the last recall time of song B is twenty minutes ago, it is determined that song a is a popular recommendation song currently being compared with hot, and song a is ranked in front of song B.
When the first recall content obtained from the first recall pool by the recommendation system has a large recall amount, the candidate recall content meeting the conditions determined according to the historical recall record has a large number, and in order to ensure the timeliness and the accuracy of the recall content in the second recall pool, the sorted candidate recall content also needs to be further screened, and for example, the recall content with the top rank in the candidate recall content is screened according to the recall proportion set by the recommendation system. Recall content in the second recall pool is available at a recall ratio of 20% of the candidate recall content as set by the recommendation system.
Taking the recommended songs as an example, when the candidate recall songs are ranked by the recommendation system, the number of the candidate songs reaches 1000, and in order to ensure timeliness of the heat of the recall songs in the second recall pool, the top 20% of the 1000 songs are selected as recall contents in the second recall pool, that is, the top 200 songs are selected as recall songs to be updated.
And taking the content obtained after screening the candidate recall content as the recall content in the second recall pool, and updating the second recall pool according to the screening configuration parameters and the updated historical recall record.
And if the configuration parameters are set to be updated at regular intervals, only recalling contents meeting the conditions in the last thirty minutes are stored in the second recall pool, namely history records exceeding thirty minutes are updated, and the recalling contents in the second recall pool are updated.
Step 607, obtaining a second recall content for a second recall amount from a front portion of the second recall pool based on the historical number of recalls and the most recent recall time.
After the second recall pool is updated, the latest popular recall content is ensured to be stored in the second recall pool, and the recall content in the second recall pool is sorted in descending order based on the historical recall times and the latest recall time. And when the first recall content acquired by the recommending system for the cold-start user does not exceed the recall quantity threshold, determining a second recall quantity according to the recall quantity threshold and the difference value of the first recall quantity corresponding to the first recall content, and selecting the second recall content of the second recall quantity from the second recall pool according to the sequence from front to back.
In the embodiment of the application, the server updates the corresponding historical recall record according to the target recall content recommended to the user, selects the candidate recall content according to the updated historical recall record, and sorts and screens the candidate recall content based on the set configuration parameters and the updated historical recall record, so that the second recall pool is updated.
For a cold-start user, when the first recall content acquired by the cold-start user does not exceed the recall amount threshold value, the recommendation system selects the second recall content of the second recall amount from the second recall pool to supplement, so as to achieve the condition of recommending the content to the user, provide the bibliographic data under the condition that the recall content is insufficient, push the latest popular recall content to the user, ensure the timeliness of the recommended content and provide diversified recommended content to the user.
Referring to fig. 7, a schematic diagram of an implementation of a content recommendation method according to an exemplary embodiment of the present application is shown, and this embodiment takes the method applied to the server shown in fig. 1 as an example for description.
As shown in fig. 7, when a user needs to obtain recommended content, a content recommendation request 701 is sent to a server through a trigger terminal. The server is connected with the terminal through the gateway device, and when the server receives a content recommendation request 701 sent by the terminal, the server acquires user information contained in the corresponding content recommendation request 701.
Further, the server acquires the first recall content from the first recall pool 703 based on the acquired user information and the configuration information 702 of the recommendation system, and for the cold start user, the server cannot acquire enough user information and cannot acquire the first recall content of the recall volume threshold from the first recall pool 703, at this time, the recommendation system determines that the second recall content of the second recall volume needs to be acquired from the second recall pool 704 according to the set recall volume threshold and the difference value of the first recall content, and the acquired second recall content is the latest popular recall content stored in the second recall pool 704. And determining the union of the first recall content and the second recall content as third recall content, and filtering and screening the third recall content according to the historical recommendation record 705 corresponding to the user, so as to ensure that the screened target recall content does not contain the content in the historical recommendation record, and ensure the novelty of the recommended content.
And when the target recall content is determined, sequencing the target recall content, determining a recommendation priority in a mode that the first recall content is arranged in front and the second recall content is arranged behind, pushing the content to the user, and then obtaining a recommendation result 706 by the user.
Further, the recommendation system updates the historical recommendation record 705 and the second recall pool 704 according to the determined target recall content, and the recall content of the second recall pool 704 is updated according to the filtering configuration parameters and the updated historical recall record, the configuration process can be updated by setting the configuration parameters, such as updating the second recall pool 704 by setting the parameter percentage (P), the latest recall time (M) and the minimum number of recalls (N), such as setting the parameters P-20, M-30 and N-100 of the recommendation system, which means that the former 20% of the target recall content with the number of recalls greater than or equal to 100 in 30 minutes can be used as the second recall pool 704, and the recall content in the second recall pool 704 is sorted according to the number of recalls and the latest recall time, i.e. the content with the number of recalls is arranged at the top, and when the recall content with the same number of recalls, the ranking with the more recent recall time is determined to be before. When the first recall content is insufficient, a second recall content of a second recall amount is selected from the second recall pool 704 in a forward-to-backward order based on the determined second recall amount.
In addition, because the created second recall pool 704 is stored with the hot recall content updated and created in real time, the created second recall pool 704 may also be used as a multi-service sharing recommendation platform, that is, under the condition of a multi-service sharing unified recommendation platform, the music platform a has a corresponding second recall pool, a current hot recall song is stored in the second recall pool, after the second recall pool is used as a sharing recommendation platform, the music platform B may also obtain the hot recall music from the second recall pool 704 in the music platform a as required, and the second recall pool 704 may update the hot recall music in real time according to the music platform B, so that the data interaction between different platforms is also achieved by reducing the large data repetition operation in a manner of sharing the recommendation platform.
Fig. 8 is a block diagram of a content recommendation apparatus according to an exemplary embodiment of the present application. The device is used for a server, and the device comprises:
a receiving module 801, configured to receive a content recommendation request sent by a terminal, where the content recommendation request includes user information;
a first recall module 802, configured to obtain first recall content from a first recall pool based on the user information, where the first recall pool includes full-volume recall content;
a second recall module 803, configured to, in response to the recall amount of the first recall content being less than the recall amount threshold, obtain second recall content from a second recall pool, where the recall content in the second recall pool is determined to be obtained based on the historical recall record of the recall content in the first recall pool;
a pushing module 804, configured to push the targeted recall content to the terminal based on the first recall content and the second recall content.
Optionally, the apparatus further comprises:
an obtaining module, configured to obtain the historical recall record of each piece of recall content in the first recall pool;
and the screening module is used for screening the recall content in the second recall pool from the first recall pool based on the historical recall record and the screening configuration parameters.
Optionally, the screening configuration parameters include a recall time period, a minimum number of recalls, and a recall ratio;
the screening module includes:
a first determining unit, configured to determine, based on the historical recall record, candidate recall content in the first recall pool, where the number of recalls in the recall time period is greater than or equal to the minimum number of recalls;
a ranking unit configured to rank the candidate recall content based on the number of recalls;
and the screening unit is used for screening the recall contents in the second recall pool from the sorted candidate recall contents based on the recall proportion.
Optionally, the apparatus further comprises:
a first update module for updating the historical recall time and the recall times of the targeted recall content in the first recall pool;
and the second updating module is used for responding to the condition that a recall pool is updated, and updating the second recall pool based on the screening configuration parameters and the updated historical recall record.
Optionally, the pushing module 804 includes:
a second determination unit configured to determine a union of the first recall content and the second recall content as a third recall content;
a third determining unit, configured to determine the target recall content based on the third recall content and a history recommendation record corresponding to the user information, where the target recall content does not include recall content in the history recommendation record;
and the pushing unit is used for pushing the target recall content to the terminal.
Optionally, the pushing unit is configured to:
determining a recommendation priority of each piece of the target recall content;
and sequencing the target recall content based on the recommendation priority, and pushing the sequenced target recall content to the terminal.
Optionally, the pushing unit is further configured to:
determining the recommendation priority based on the source of each of the targeted recall content, wherein the recommendation priority of the targeted recall content from the first recall pool is higher than the recommendation priority of the targeted recall content from the second recall pool;
and/or the presence of a gas in the gas,
determining the recommendation priority based on a historical number of recalls of each of the targeted recall content.
Optionally, the apparatus further comprises:
and the third updating module is used for updating the historical recommendation record corresponding to the user information based on the target recall content.
Optionally, the second recall module 803 includes:
a fourth determining unit, configured to determine a second recall amount based on the first recall amount and the recall amount threshold, where the second recall amount is in a negative correlation with the first recall amount;
a first obtaining unit, configured to obtain the second recall content of the second recall amount from the second recall pool.
Optionally, the recall contents in the second recall pool are sorted in descending order based on the historical recall times and the latest recall time;
the first obtaining unit is further configured to:
obtaining the second recall content for the second recall amount from a front of the second recall pool.
To sum up, in the method provided by the embodiment of the present application, the server obtains the user information included in the content recommendation request according to the obtained content recommendation request sent by the terminal, and further obtains the first recall content from the first recall pool according to the user information; in addition, considering that the cold start stage is only based on the user information, the sufficient content is not recalled from the first recall pool, the first recall content and the second recall content are pushed to the terminal by acquiring the second recall content from the second recall pool, and the sufficient recommended content is pushed to the user; in addition, the second recall content is determined and obtained based on the historical recall record of the recall content in the first recall pool, so by adopting the scheme provided by the embodiment of the application, the content recall quality can be improved and the probability of using the recall content by a subsequent user can be improved on the premise of ensuring that enough content is recalled.
In the embodiment of the application, when the server is insufficient to obtain enough first recall content from the first recall pool based on the obtained user information data, the second recall content is obtained from the second recall pool for parallel supplement, and the union of the first recall content and the second recall content is determined as the third recall content, so that the recall volume of the recall content reaches a recall volume threshold value, and the provision of the bibliographic data is performed under the condition that the recall content is insufficient;
furthermore, the third recall content is screened according to the historical recommendation record and sorted according to the recommendation priority, so that personalized target recall content is recommended to the user, and the phenomenon that the experience of the user is reduced due to the simplification of the recommended content is avoided.
In the embodiment of the application, the server updates the corresponding historical recall record according to the target recall content recommended to the user, selects the candidate recall content according to the updated historical recall record, and sorts and screens the candidate recall content based on the set configuration parameters and the updated historical recall record, so that the second recall pool is updated.
For a cold-start user, when the first recall content acquired by the cold-start user does not exceed the recall amount threshold value, the recommendation system selects the second recall content of the second recall amount from the second recall pool to supplement, so as to achieve the condition of recommending the content to the user, provide the bibliographic data under the condition that the recall content is insufficient, push the latest popular recall content to the user, ensure the timeliness of the recommended content and provide diversified recommended content to the user.
Referring to fig. 9, a block diagram of a server according to an embodiment of the present application is shown, where the server may be used to implement the content recommendation method executed by the server according to the embodiment. The server 900 includes a Central Processing Unit (CPU) 901, a system Memory 904 including a Random Access Memory (RAM) 902 and a Read-Only Memory (ROM) 903, and a system bus 905 connecting the system Memory 904 and the Central Processing unit 901. The server 900 also includes a basic Input/Output system (I/O) 906, which facilitates the transfer of information between devices within the computer, and a mass storage device 907 for storing an operating system 913, application programs 914, and other program modules 915.
The basic input/output system 906 includes a display 908 for displaying information and an input device 909 such as a mouse, keyboard, etc. for user input of information. Wherein the display 908 and the input device 909 are connected to the central processing unit 901 through an input/output controller 910 connected to the system bus 905. The basic input/output system 906 may also include an input/output controller 910 for receiving and processing input from a number of other devices, such as a keyboard, mouse, or electronic stylus. Similarly, an input/output controller 910 also provides output to a display screen, a printer, or other type of output device.
The mass storage device 907 is connected to the central processing unit 901 through a mass storage controller (not shown) connected to the system bus 905. The mass storage device 907 and its associated computer-readable media provide non-volatile storage for the server 900. That is, the mass storage device 907 may include a computer-readable medium (not shown) such as a hard disk or CD-ROM (Compact disk Read-Only Memory) drive.
Without loss of generality, the computer-readable media may comprise computer storage media and communication media. Computer storage media includes volatile and nonvolatile, removable and non-removable media implemented in any method or technology for storage of information such as computer readable instructions, data structures, program modules or other data. Computer storage media includes RAM, ROM, EPROM (Erasable Programmable Read-Only Memory), EEPROM (Electrically Erasable Programmable Read-Only Memory), flash Memory or other solid state Memory device, CD-ROM, DVD (Digital Video Disc) or other optical storage, magnetic cassettes, magnetic tape, magnetic disk storage or other magnetic storage devices. Of course, those skilled in the art will appreciate that the computer storage media is not limited to the foregoing. The system memory 904 and mass storage device 907 described above may be collectively referred to as memory.
The server 900 may also operate as a remote computer connected to a network via a network, such as the internet, according to various embodiments of the present application. That is, the server 900 may be connected to the network 912 through the network interface unit 911 coupled to the system bus 905, or the network interface unit 911 may be used to connect to other types of networks or remote computer systems (not shown).
The memory also includes one or more programs stored in the memory and configured to be executed by the one or more central processing units 901.
The present application further provides a computer-readable storage medium having at least one instruction, at least one program, a set of codes, or a set of instructions stored therein, which is loaded and executed by a processor to implement the content recommendation method provided by any of the above exemplary embodiments.
Embodiments of the present application provide a computer program product or computer program comprising computer instructions stored in a computer readable storage medium. The processor of the computer device reads the computer instructions from the computer-readable storage medium, and the processor executes the computer instructions to cause the computer device to execute the content recommendation method provided in the above-described alternative implementation.
It will be understood by those skilled in the art that all or part of the steps for implementing the above embodiments may be implemented by hardware, or may be implemented by a program instructing relevant hardware, where the program may be stored in a computer-readable storage medium, and the above-mentioned storage medium may be a read-only memory, a magnetic disk or an optical disk, etc.
The above description is only exemplary of the present application and should not be taken as limiting, as any modification, equivalent replacement, or improvement made within the spirit and principle of the present application should be included in the protection scope of the present application.
Claims (13)
1. A method for recommending content, the method comprising:
receiving a content recommendation request sent by a terminal, wherein the content recommendation request comprises user information;
acquiring first recall content from a first recall pool based on the user information, wherein the first recall pool comprises full recall content;
in response to the recall amount of the first recall content being less than a recall amount threshold, obtaining second recall content from a second recall pool, the recall content in the second recall pool determined to be obtained based on historical recall records of recall content in the first recall pool;
and pushing target recall content to the terminal based on the first recall content and the second recall content.
2. The method of claim 1, further comprising:
acquiring the historical recall record of each piece of recall content in the first recall pool;
and screening the recall content in the second recall pool from the first recall pool based on the historical recall record and screening configuration parameters.
3. The method of claim 2, wherein the screening configuration parameters include a recall period, a minimum number of recalls, and a recall proportion;
the filtering recall content in the second recall pool from the first recall pool based on the historical recall records and filtering configuration parameters, comprising:
determining candidate recall content in the first recall pool with a recall number greater than or equal to the minimum recall number within the recall period based on the historical recall record;
ranking the candidate recall content based on the number of recalls;
and screening the recall contents in the second recall pool from the sorted candidate recall contents based on the recall proportion.
4. The method of claim 2, wherein after the pushing of the targeted recall content to the terminal based on the first recall content and the second recall content, the method further comprises:
updating the historical recall times and the number of recalls of the targeted recall content in the first recall pool;
in response to a recall pool update condition being met, updating the second recall pool based on the screening configuration parameters and the updated historical recall records.
5. The method according to any one of claims 1 to 4, wherein the pushing targeted recall content to the terminal based on the first recall content and the second recall content comprises:
determining a union of the first recall content and the second recall content as a third recall content;
determining the target recall content based on the third recall content and a historical recommendation record corresponding to the user information, wherein the target recall content does not include recall content in the historical recommendation record;
and pushing the target recall content to the terminal.
6. The method of claim 5, wherein the pushing the targeted recall content to the terminal comprises:
determining a recommendation priority of each piece of the target recall content;
and sequencing the target recall content based on the recommendation priority, and pushing the sequenced target recall content to the terminal.
7. The method of claim 6, wherein said determining a recommendation priority for each of said targeted recall content comprises:
determining the recommendation priority based on the source of each of the targeted recall content, wherein the recommendation priority of the targeted recall content from the first recall pool is higher than the recommendation priority of the targeted recall content from the second recall pool;
and/or the presence of a gas in the gas,
determining the recommendation priority based on a historical number of recalls of each of the targeted recall content.
8. The method of claim 5, wherein after the pushing of the targeted recall content to the terminal, the method further comprises:
and updating the historical recommendation record corresponding to the user information based on the target recall content.
9. The method of any of claims 1 to 4, wherein said retrieving second recall content from a second recall pool comprises:
determining a second recall amount based on the first recall amount and the recall amount threshold, the second recall amount being inversely related to the first recall amount;
obtaining the second recall content for the second recall amount from the second recall pool.
10. The method of claim 9, wherein the recall content in the second recall pool is sorted in descending order based on historical recall times and most recent recall times;
the obtaining the second recall content of the second recall amount from the second recall pool comprises:
obtaining the second recall content for the second recall amount from a front of the second recall pool.
11. A content recommendation apparatus, the apparatus comprising:
the system comprises a receiving module, a sending module and a receiving module, wherein the receiving module is used for receiving a content recommendation request sent by a terminal, and the content recommendation request comprises user information;
the first recall module is used for acquiring first recall content from a first recall pool based on the user information, wherein the first recall pool comprises full recall content;
a second recall module, configured to, in response to a recall amount of the first recall content being less than a recall amount threshold, obtain second recall content from a second recall pool, where the recall content in the second recall pool is determined to be obtained based on a historical recall record of recall content in the first recall pool;
and the pushing module is used for pushing the target recall content to the terminal based on the first recall content and the second recall content.
12. A server, characterized in that the server comprises a processor and a memory, in which at least one instruction, at least one program, a set of codes, or a set of instructions is stored, which is loaded and executed by the processor to implement the content recommendation method according to any one of claims 1 to 10.
13. A computer-readable storage medium, having stored therein at least one instruction, at least one program, a set of codes, or a set of instructions, which is loaded and executed by a processor to implement the content recommendation method of any of claims 1 to 10.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110340317.XA CN113010790A (en) | 2021-03-30 | 2021-03-30 | Content recommendation method, device, server and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110340317.XA CN113010790A (en) | 2021-03-30 | 2021-03-30 | Content recommendation method, device, server and storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN113010790A true CN113010790A (en) | 2021-06-22 |
Family
ID=76409263
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110340317.XA Pending CN113010790A (en) | 2021-03-30 | 2021-03-30 | Content recommendation method, device, server and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113010790A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113689232A (en) * | 2021-07-27 | 2021-11-23 | 阿里巴巴（中国）有限公司 | Method and device for detecting crowd recall service and electronic equipment |
| CN114785852A (en) * | 2022-04-07 | 2022-07-22 | 北京达佳互联信息技术有限公司 | Push content determining method and device, electronic equipment and storage medium |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101370055A (en) * | 2008-09-24 | 2009-02-18 | 中国电信股份有限公司 | Propelling movement method, platform and system for recommending information of personalized ring back tone |
| CN102402625A (en) * | 2011-12-28 | 2012-04-04 | 深圳市五巨科技有限公司 | Method and system for recommending music |
| CN110543598A (en) * | 2019-09-06 | 2019-12-06 | 腾讯科技（深圳）有限公司 | information recommendation method and device and terminal |
| CN111859126A (en) * | 2020-07-09 | 2020-10-30 | 有半岛(北京)信息科技有限公司 | Recommended item determination method, device, equipment and storage medium |
| CN111931055A (en) * | 2020-08-14 | 2020-11-13 | 工银科技有限公司 | Object recommendation method, object recommendation device and electronic equipment |
-
2021
- 2021-03-30 CN CN202110340317.XA patent/CN113010790A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101370055A (en) * | 2008-09-24 | 2009-02-18 | 中国电信股份有限公司 | Propelling movement method, platform and system for recommending information of personalized ring back tone |
| CN102402625A (en) * | 2011-12-28 | 2012-04-04 | 深圳市五巨科技有限公司 | Method and system for recommending music |
| CN110543598A (en) * | 2019-09-06 | 2019-12-06 | 腾讯科技（深圳）有限公司 | information recommendation method and device and terminal |
| CN111859126A (en) * | 2020-07-09 | 2020-10-30 | 有半岛(北京)信息科技有限公司 | Recommended item determination method, device, equipment and storage medium |
| CN111931055A (en) * | 2020-08-14 | 2020-11-13 | 工银科技有限公司 | Object recommendation method, object recommendation device and electronic equipment |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113689232A (en) * | 2021-07-27 | 2021-11-23 | 阿里巴巴（中国）有限公司 | Method and device for detecting crowd recall service and electronic equipment |
| CN113689232B (en) * | 2021-07-27 | 2023-07-28 | 阿里巴巴华北技术有限公司 | Method and device for detecting crowd recall service and electronic equipment |
| CN114785852A (en) * | 2022-04-07 | 2022-07-22 | 北京达佳互联信息技术有限公司 | Push content determining method and device, electronic equipment and storage medium |
| CN114785852B (en) * | 2022-04-07 | 2024-04-09 | 北京达佳互联信息技术有限公司 | Push content determining method and device, electronic equipment and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| RU2720899C2 (en) | Method and system for determining user-specific content proportions for recommendation | |
| US10469549B2 (en) | Device for participating in a network for sharing media consumption activity | |
| US7680959B2 (en) | P2P network for providing real time media recommendations | |
| US9552428B2 (en) | System for generating media recommendations in a distributed environment based on seed information | |
| US7970922B2 (en) | P2P real time media recommendations | |
| US10277696B2 (en) | Method and system for processing data used by creative users to create media content | |
| CN102473189B (en) | Providing in real time to the link of portion of media object during social networks upgrades | |
| CN104813256A (en) | Gathering and organizing content distributed via social media | |
| CN111432226B (en) | Live broadcast recommendation method and device, server, terminal and storage medium | |
| WO2018133725A1 (en) | Method, client, and system for output processing of media file | |
| WO2013189296A1 (en) | Method and system for processing recommended target software | |
| KR20110066168A (en) | System and method for playlist generation based on similarity data | |
| WO2013184957A1 (en) | Systems and methods of classifying content items | |
| CN111782957A (en) | Method and device for recommending content, computing equipment and storage medium | |
| US20150066897A1 (en) | Systems and methods for conveying passive interest classified media content | |
| CN104954872B (en) | video downloading method, video playing device, mobile terminal and storage medium | |
| CN113010790A (en) | Content recommendation method, device, server and storage medium | |
| US8725795B1 (en) | Content segment optimization techniques | |
| US20080307316A1 (en) | System and method for assigning user preference settings to fields in a category, particularly a media category | |
| CN111949807A (en) | Data processing method, data processing device, computer equipment and storage medium | |
| CN111626615A (en) | Method, device and computer storage medium for determining singing order score | |
| CN111683280A (en) | Video processing method and device and electronic equipment | |
| CN108134948B (en) | Television program recommendation method, device and system and readable storage medium | |
| CN115827978A (en) | Information recommendation method, device, equipment and computer readable storage medium | |
| CN114020980A (en) | Method for generating and creating content list, electronic equipment and storage medium |
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 |