CN113139086A - Information recommendation method and device - Google Patents

Information recommendation method and device Download PDF

Info

Publication number
CN113139086A
CN113139086A CN202110475167.3A CN202110475167A CN113139086A CN 113139086 A CN113139086 A CN 113139086A CN 202110475167 A CN202110475167 A CN 202110475167A CN 113139086 A CN113139086 A CN 113139086A
Authority
CN
China
Prior art keywords
recommendation
media information
user
media
information
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
Application number
CN202110475167.3A
Other languages
Chinese (zh)
Inventor
刘伟科
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Beijing Wodong Tianjun Information Technology Co Ltd
Original Assignee
Beijing Wodong Tianjun Information Technology Co Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Beijing Wodong Tianjun Information Technology Co Ltd filed Critical Beijing Wodong Tianjun Information Technology Co Ltd
Priority to CN202110475167.3A priority Critical patent/CN113139086A/en
Publication of CN113139086A publication Critical patent/CN113139086A/en
Pending legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/70—Information retrieval; Database structures therefor; File system structures therefor of video data
    • G06F16/73—Querying
    • G06F16/735—Filtering based on additional data, e.g. user or group profiles
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F21/00—Security arrangements for protecting computers, components thereof, programs or data against unauthorised activity
    • G06F21/60—Protecting data
    • G06F21/62—Protecting access to data via a platform, e.g. using keys or access control rules
    • G06F21/6218—Protecting access to data via a platform, e.g. using keys or access control rules to a system of files or objects, e.g. local or distributed file system or database
    • G06F21/6245—Protecting personal data, e.g. for financial or medical purposes

Abstract

The disclosure provides an information recommendation method and device. The information recommendation method comprises the following steps: responding to information recommendation triggering operation performed by a user, and determining a media information recommendation mode according to a recommendation count value; if the media information recommendation mode is the first recommendation mode, selecting a piece of media information from an original media library by using an original recommendation strategy and recommending the media information to a user, and adding 1 to a recommendation count value according to a preset condition so as to update the recommendation count value; if the media information recommendation mode is the second recommendation mode, selecting a target record associated with the user from the repeated recommendation list; and pushing the media information associated with the media information identification in the target record to the user, and resetting the recommendation count value.

Description

Information recommendation method and device
Technical Field
The present disclosure relates to the field of information processing, and in particular, to an information recommendation method and apparatus.
Background
In the related art, after a user logs in a video platform, the video platform selects a corresponding video from a video library according to a preset recommendation strategy to recommend the video to the user. In addition, the user adds the video information of interest to the focus list so that the video of interest can be repeatedly viewed.
Disclosure of Invention
The inventors have noted that the recommendation strategies used in the related art all follow the logic of "finding approximate videos", i.e., videos viewed by the user are not recommended repeatedly by the video platform. Under the condition that a user adds interested video information into an attention list, if the attention list has more information, the user can hardly find the interested video information quickly. Also the focus list may expose the personal privacy of the user.
Accordingly, the information recommendation scheme is provided, and the media information interested by the user can be repeatedly recommended to the user under the condition of protecting the personal privacy of the user.
According to a first aspect of the embodiments of the present disclosure, there is provided an information recommendation method, including: responding to information recommendation triggering operation performed by a user, and determining a media information recommendation mode according to a recommendation count value; if the media information recommendation mode is a first recommendation mode, selecting a piece of media information from an original media library by using an original recommendation strategy and recommending the media information to the user, and adding 1 to the recommendation count value according to a preset condition so as to update the recommendation count value; if the media information recommendation mode is a second recommendation mode, selecting a target record associated with the user from a repeated recommendation list; and pushing the media information associated with the media information identifier in the target record to the user, and resetting the recommendation count value.
In some embodiments, after the media information associated with the media information identifier is pushed to the user, the number of remaining recommendations in the target record is updated; and if the residual recommendation times in the target record are 0, deleting the target record.
In some embodiments, determining a media information recommendation manner according to the recommendation count value includes: judging whether the recommended count value is smaller than a preset count threshold value or not; if the recommendation count value is smaller than the preset count threshold value, determining that the media information recommendation mode is the first recommendation mode; and if the recommendation count value is equal to the preset count threshold value, determining that the media information recommendation mode is the second recommendation mode.
In some embodiments, the preset condition comprises updating the recommendation count value by adding 1 to the recommendation count value after recommending to the user media information selected from an original media library.
In some embodiments, the preset condition further includes, after recommending the media information selected from the original media library to the user, in case that the media information selected from the original media library is of a predetermined type, adding 1 to the recommendation count value to update the recommendation count value.
In some embodiments, the type of media information associated with the media information identification is different from the predetermined type.
In some embodiments, determining a media information recommendation manner according to the recommendation count value includes: configuring a first probability for the first recommendation mode and a second probability for the second recommendation mode, wherein the sum of the first probability and the second probability is 1, and the second probability and the recommendation count value have positive correlation; selecting the first recommendation mode and the second recommendation mode according to the first probability and the second probability; and determining the media information recommendation mode according to the selection result.
In some embodiments, the above method further comprises: collecting user behaviors of the user aiming at the currently played media information; determining whether the currently played media information needs to be written into a repeated recommendation list according to the user behavior; if the currently played media information is determined to be required to be written into a repeated recommendation list, inquiring whether a record associated with the user identification of the user and the information identification of the currently played media information is included in the repeated recommendation list; and if the repeated recommendation list does not comprise a record associated with the user identifier of the user and the information identifier of the currently played media information, creating a new record in the repeated recommendation list, wherein the new record comprises the user identifier of the user, the information identifier of the currently played media information and the preset residual recommendation times.
In some embodiments, if the repeated recommendation list includes a record associated with the user identifier of the user and the information identifier of the currently played media information, the remaining recommendation times are updated to a preset remaining recommendation times in the record associated with the user identifier of the user and the information identifier of the currently played media information.
In some embodiments, determining whether the currently played media information needs to be written into a duplicate recommendation list based on the user behavior comprises: detecting whether the user behaviors comprise behaviors which are not interested in media information; if the user behavior comprises a behavior which is not interested in the media information, determining that the currently played media information does not need to be written into a repeated recommendation list; if the user behaviors do not include behaviors which are not interested in the media information, further detecting whether the user behaviors include behaviors which are interested in the media information; and if the user behaviors include the behaviors interested in the media information, determining that the currently played media information needs to be written into a repeated recommendation list.
In some embodiments, determining whether the currently played media information needs to be written into a duplicate recommendation list based on the user behavior comprises: obtaining a score corresponding to the user behavior by using a user interest behavior library, and determining a summary value according to the obtained score, wherein the user interest behavior library is constructed by using a historical behavior library, and the user interest behavior library comprises behaviors which are interested in media and corresponding scores, and behaviors which are not interested in the media and corresponding scores; and determining whether the currently played media information needs to be written into a repeated recommendation list according to the summary value.
In some embodiments, determining whether the currently played media information needs to be written into the repeated recommendation list according to the summary value comprises: and if the summary value associated with the behavior interested in the media, which is included in the summary value, is greater than a preset summary threshold value, determining that the currently played media information needs to be written into a repeated recommendation list.
In some embodiments, determining whether the currently played media information needs to be written into the repeated recommendation list according to the summary value comprises: and if the ratio of the summary value associated with the behavior interested in the media in the summary value to the summary value associated with the behavior not interested in the media in the summary value is greater than a preset proportion threshold, determining that the currently played media information needs to be written into a repeated recommendation list.
In some embodiments, after pushing the media information associated with the media information identification to the user, collecting the user's viewing behavior; and updating the scores of the corresponding behaviors in the user interest behavior library of the user and the scores of the corresponding behaviors in the historical behavior library according to the watching behaviors of the user.
According to a second aspect of the embodiments of the present disclosure, there is provided an information recommendation apparatus including: the first processing module is configured to respond to information recommendation triggering operation performed by a user and determine a media information recommendation mode according to a recommendation count value; the second processing module is configured to select a piece of media information from an original media library by using an original recommendation strategy and recommend the piece of media information to the user if the media information recommendation mode is the first recommendation mode, and add 1 to the recommendation count value according to a preset condition so as to update the recommendation count value; and if the media information recommendation mode is a second recommendation mode, selecting a target record associated with the user from a repeated recommendation list, pushing media information associated with the media information identifier in the target record to the user, and resetting the recommendation count value.
According to a third aspect of the embodiments of the present disclosure, there is provided an information recommendation apparatus including: a memory configured to store instructions; a processor coupled to the memory, the processor configured to perform a method implementing any of the embodiments described above based on instructions stored by the memory.
According to a fourth aspect of the embodiments of the present disclosure, a computer-readable storage medium is provided, in which computer instructions are stored, and when executed by a processor, the computer-readable storage medium implements the method according to any of the embodiments described above.
Other features of the present disclosure and advantages thereof will become apparent from the following detailed description of exemplary embodiments thereof, which proceeds with reference to the accompanying drawings.
Drawings
In order to more clearly illustrate the embodiments of the present disclosure or the technical solutions in the prior art, the drawings needed to be used in the description of the embodiments or the prior art will be briefly introduced below, it is obvious that the drawings in the following description are only some embodiments of the present disclosure, and for those skilled in the art, other drawings can be obtained according to the drawings without inventive exercise.
FIG. 1 is a flowchart illustrating a duplicate recommendation list update method according to an embodiment of the present disclosure;
FIG. 2 is a flowchart illustrating an information recommendation method according to an embodiment of the disclosure;
fig. 3 is a schematic structural diagram of an information recommendation device according to an embodiment of the present disclosure;
fig. 4 is a schematic structural diagram of an information recommendation device according to another embodiment of the present disclosure;
fig. 5 is a schematic structural diagram of an information recommendation device according to yet another embodiment of the present disclosure.
Detailed Description
The technical solutions in the embodiments of the present disclosure will be clearly and completely described below with reference to the drawings in the embodiments of the present disclosure, and it is obvious that the described embodiments are only a part of the embodiments of the present disclosure, and not all of the embodiments. The following description of at least one exemplary embodiment is merely illustrative in nature and is in no way intended to limit the disclosure, its application, or uses. All other embodiments, which can be derived by a person skilled in the art from the embodiments disclosed herein without making any creative effort, shall fall within the protection scope of the present disclosure.
The relative arrangement of the components and steps, the numerical expressions, and numerical values set forth in these embodiments do not limit the scope of the present disclosure unless specifically stated otherwise.
Meanwhile, it should be understood that the sizes of the respective portions shown in the drawings are not drawn in an actual proportional relationship for the convenience of description.
Techniques, methods, and apparatus known to those of ordinary skill in the relevant art may not be discussed in detail but are intended to be part of the specification where appropriate.
In all examples shown and discussed herein, any particular value should be construed as merely illustrative, and not limiting. Thus, other examples of the exemplary embodiments may have different values.
It should be noted that: like reference numbers and letters refer to like items in the following figures, and thus, once an item is defined in one figure, further discussion thereof is not required in subsequent figures.
Fig. 1 is a flowchart illustrating a repetitive recommendation list updating method according to an embodiment of the present disclosure. In some embodiments, the following repeated recommendation list updating method steps are performed by the information recommendation device.
In step 101, user behavior of a user with respect to currently played media information is collected.
In step 102, it is determined whether the currently played media information needs to be written into the repeated recommendation list according to the user behavior.
In some embodiments, the historical behavior library is constructed by aggregating behaviors that different users are interested in media information when viewing the media information and behaviors that are not interested in the media information. For example, actions that a user is interested in media information include: focusing on media information, viewing in cycles, viewing other media information of the same anchor, viewing comments, pausing, viewing homogeneous media information, searching for media information, forwarding media information, adjusting volume, body motion, and the like. The behavior of the user not interested in the media information comprises: the method comprises the steps of not seeing the video, quickly swiping, clicking a dislike button, swiping after seeing the same type of media information, not opening the media information and the like.
In some embodiments, an initial score may be set for each behavior. The initial scores for each behavior may be the same or different. The historical behavior library is shown in table 1.
Figure BDA0003046725860000061
TABLE 1
In some embodiments, the user interest behavior library for each user is built by replicating a historical behavior library. The user interest behavior library is shown in table 2.
Figure BDA0003046725860000062
Figure BDA0003046725860000071
TABLE 2
In some embodiments, a user interest behavior library is utilized to detect whether behaviors that are not interested in media information are included in user behaviors. And if the user behaviors comprise behaviors which are not interested in the media information, determining that the currently played media information does not need to be written into the repeated recommendation list. And if the user behaviors do not include the behaviors which are not interested in the media information, further detecting whether the user behaviors include the behaviors which are interested in the media information. And if the user behaviors comprise the behaviors interested in the media information, determining that the currently played media information needs to be written into a repeated recommendation list.
That is, as long as there is an action of no interest, the currently played media information is not written in the repeated recommendation list. Or, if only the behavior of interest exists, the currently played media information is written into the repeated recommendation list.
In other embodiments, a score corresponding to the user behavior is obtained by using the user interest behavior library, and a summary value is determined according to the obtained score. And determining whether the currently played media information needs to be written into the repeated recommendation list according to the summary value.
For example, if the summary value associated with the behavior interested in the media included in the summary value is greater than the preset summary threshold, it is determined that the currently played media information needs to be written into the repeated recommendation list.
If the initial summary value of the user a is 0, the user a has the behaviors of playing circularly and adjusting the volume when watching the video 1. From table 2 above, the sum of user a is added to 20. If user a clicks on the dislike button behavior while watching video 1, the aggregate value for user a is subtracted by 10. And if the final summary value of the user A is larger than the preset summary threshold value, determining to write the information of the video 1 into the repeated recommendation list.
For another example, if a ratio of the summary value associated with the behavior interested in the media included in the summary value to the summary value associated with the behavior not interested in the media included in the summary value is greater than a preset ratio threshold, it is determined that the currently played media information needs to be written into the repeated recommendation list.
If the score corresponding to the interesting behavior is 70 and the score corresponding to the uninteresting behavior is 20 in the process of watching the video 1, the ratio of 70/20 is greater than the preset ratio threshold, and it is determined that the information of the video 1 is written into the repeated recommendation list.
In step 103, if it is determined that the currently played media information needs to be written into the repeated recommendation list, whether a record associated with the user identifier of the user and the information identifier of the currently played media information is included in the repeated recommendation list is queried.
In step 104, if the repeated recommendation list does not include a record associated with the user identifier of the user and the information identifier of the currently played media information, a new record is created in the repeated recommendation list, where the new record includes the user identifier of the user, the information identifier of the currently played media information, and the preset remaining recommendation times.
For example, it is determined that the video 3 needs to be written in the repeated recommendation list according to the behavior of the user a when viewing the video 3. If the duplicate recommendation list does not include the record associated with the user a and the video 3, a new record is created in the duplicate recommendation list, where the new record includes the user identifier of the user a, the information identifier of the video 3, and the preset remaining recommendation times, as shown in table 3. For example, the number of remaining recommendations is preset to 5.
Figure BDA0003046725860000081
TABLE 3
In step 105, if the repeated recommendation list includes a record associated with the user identifier of the user and the information identifier of the currently played media information, the remaining recommendation times is updated to the preset remaining recommendation times in the record associated with the user identifier of the user and the information identifier of the currently played media information.
For example, it is determined that video 1 needs to be written in the duplicate recommendation list according to the behavior of user a when viewing video 1. If the duplicate recommendation list includes records associated with user a and video 1, as shown in table 3, the number of remaining recommendations in the records associated with user a and video 1 is revised again to 5 times in the duplicate recommendation list, as shown in table 4.
Figure BDA0003046725860000091
TABLE 4
Fig. 2 is a schematic flow chart of an information recommendation method according to an embodiment of the present disclosure. In some embodiments, the following information recommendation method steps are performed by an information recommendation device.
In step 201, in response to an information recommendation triggering operation performed by a user, a media information recommendation mode is determined according to a recommendation count value.
In some embodiments, the user operates through a gliding interface or on a waterfall-type page to perform an information recommendation triggering operation.
It should be noted that the recommendation count value corresponds to the number of times the user is continuously recommended with the media information in the original media library. And if the repeated recommendation list is used for recommending the media information to the user, resetting the recommendation count value to be 0.
In some embodiments, the recommended count value is determined by determining whether the recommended count value is less than a preset count threshold. And if the recommendation count value is smaller than the preset count threshold value, determining that the media information recommendation mode is the first recommendation mode. And if the recommendation count value is equal to the preset count threshold value, determining that the media information recommendation mode is the second recommendation mode.
For example, if the preset count threshold is 3, after continuously recommending 3 pieces of media information from the original media library, recommending the media information that the user wants to be repeatedly recommended to the user by using the repeated recommendation list. Therefore, the media information which the user wants to recommend repeatedly can appear more naturally, and the preference privacy of the user can not be leaked.
In other embodiments, a first probability is configured for a first recommendation method for recommending media information from an original media library, a second probability is configured for a second recommendation method for recommending media information by using a repeated recommendation list, the sum of the first probability and the second probability is 1, and the second probability and a recommendation count value have a positive correlation. And selecting the first recommendation mode and the second recommendation mode according to the first probability and the second probability, and determining the media information recommendation mode according to the selection result.
For example, if the current recommendation count value is 0, the probability of being configured for the first recommendation method is 0.9, and the probability of being configured for the second recommendation method is 0.1, and the selection is performed between the first recommendation method and the second recommendation method according to the configuration. And if the selection result is the first recommendation mode, recommending the media information by using the first recommendation mode, and setting the current recommendation count value to be 1. And if the user carries out the triggering operation again, the probability configured for the first recommendation mode is 0.8, the probability configured for the second recommendation mode is 0.2, and the user selects from the first recommendation mode and the second recommendation mode according to the configuration. By analogy, if the second recommendation method is not selected, the probability of the second recommendation method increases according to the increase of the recommendation count value, for example, 0.1, 0.2, 0.3, and the like, and if the probability of the second recommendation method is 1, the second recommendation method is selected certainly. Through the processing, the media information which the user wants to recommend repeatedly appears randomly, so that the preference privacy of the user is not leaked.
In step 202, if the media information recommendation mode is the first recommendation mode, a piece of media information is selected from the original media library by using the original recommendation strategy and recommended to the user, and the recommendation count value is increased by 1 according to a preset condition so as to update the recommendation count value.
In some embodiments, the preset condition includes updating a recommendation count value after recommending to the user media information selected from the original media library. That is, in the first updating mode, after recommending the media information selected from the original media library to the user, the current recommendation count value is increased by 1 so as to update the recommendation count value.
In other embodiments, the predetermined condition includes updating the recommendation count value in the event that the media information selected from the original media library is of a predetermined type after the media information selected from the original media library is recommended to the user. That is, in the second updating mode, if the media information selected from the original media library is of a predetermined type, the current recommendation count value is increased by 1 so as to update the recommendation count value. Conversely, if the media information selected from the original media library is not of the predetermined type, the recommendation count value is not updated.
It should be noted that, since it is not the inventive point of the present disclosure to recommend media information from the original media library by using the original recommendation policy, the description is not made here.
In step 203, if the media information recommendation method is the second recommendation method, a target record associated with the user is selected from the repeated recommendation list.
In some embodiments, if multiple target records associated with the user are included in the repeated recommendation list, one target record may be randomly selected.
In other embodiments, if the recommendation count value is updated only for the predetermined type of media information in the second updating manner, in a case that the repeated recommendation list includes a plurality of target records associated with the user, the type of the media information associated with the media information identifier in the selected target record is different from the predetermined type.
For example, after 5 programming videos are recommended to the user through the original media library, a music video is recommended to the user through the repeated recommendation list, and therefore user experience can be effectively improved.
In step 204, the media information associated with the media information identifier in the target record is pushed to the user and the recommendation count value is reset.
For example, by resetting the recommended count value so as to set the recommended count value to 0.
In some embodiments, the remaining number of recommendations in the target record is updated after the media information associated with the media information identification is pushed to the user. And if the residual recommendation times in the target record are 0, deleting the target record.
For example, if the number of remaining recommendations in the target record is 3, after pushing the media information associated with the media information identifier in the target record to the user, the number of remaining recommendations in the target record is updated to 2, which indicates that the media information can be recommended for 2 times.
For another example, if the number of remaining recommendations in the target record is 1, the number of remaining recommendations in the target record is updated to 0 after media information associated with the media information identifier in the target record is pushed to the user. Since the current remaining recommendation number in the target record is 0, it indicates that the number of times of repeatedly recommending the video information to the user has reached the predetermined number of times. In which case the target record is deleted from the duplicate recommendation list.
In some embodiments, after the media information associated with the media information identifier is pushed to the user, the viewing behavior of the user is collected, and the score of the corresponding behavior in the user interest behavior library of the user and the score of the corresponding behavior in the historical behavior library are updated according to the viewing behavior of the user.
For example, user a has behaviors of viewing comments, adjusting volume, and clicking a dislike button while watching video 1. In this case, 1 is added to the scores corresponding to the behavior of viewing comments, adjusting the volume, and clicking a dislike button, as shown in table 5.
Figure BDA0003046725860000121
FIG. 5
In addition, the historical behavior library is updated correspondingly according to the behavior of the user A in the process of watching the video 1. It should be noted here that the corresponding score increment of the same behavior in the historical behavior library is smaller than the corresponding score increment in the user interest behavior library. Thereby ensuring that overall data fluctuations are small.
In the information recommendation method provided by the above embodiment of the present disclosure, the media information that the user is interested in is repeatedly recommended to the user by means of the repeated recommendation list, and the user does not need to pay attention to the interested media information, so that the personal privacy of the user can be effectively protected.
Fig. 3 is a schematic structural diagram of an information recommendation device according to an embodiment of the present disclosure. As shown in fig. 3, the information recommendation apparatus includes a first processing module 31 and a second processing module 32.
The first processing module 31 is configured to determine a media information recommendation mode according to a recommendation count value in response to an information recommendation triggering operation performed by a user.
In some embodiments, the user operates through a gliding interface or on a waterfall-type page to perform an information recommendation triggering operation.
It should be noted that the recommendation count value corresponds to the number of times the user is continuously recommended with the media information in the original media library. And if the repeated recommendation list is used for recommending the media information to the user, resetting the recommendation count value to be 0.
In some embodiments, the recommended count value is determined by determining whether the recommended count value is less than a preset count threshold. And if the recommendation count value is smaller than the preset count threshold value, determining that the media information recommendation mode is the first recommendation mode. And if the recommendation count value is equal to the preset count threshold value, determining that the media information recommendation mode is the second recommendation mode.
In other embodiments, a first probability is configured for a first recommendation method for recommending media information from an original media library, a second probability is configured for a second recommendation method for recommending media information by using a repeated recommendation list, the sum of the first probability and the second probability is 1, and the second probability and a recommendation count value have a positive correlation. And selecting the first recommendation mode and the second recommendation mode according to the first probability and the second probability, and determining the media information recommendation mode according to the selection result.
The second processing module 32 is configured to select a piece of media information from the original media library by using the original recommendation policy and recommend the piece of media information to the user if the media information recommendation mode is the first recommendation mode, and add 1 to the recommendation count value according to a preset condition so as to update the recommendation count value.
In some embodiments, the preset condition includes updating a recommendation count value after recommending to the user media information selected from the original media library. That is, in the first updating mode, after recommending the media information selected from the original media library to the user, the current recommendation count value is increased by 1 so as to update the recommendation count value.
In other embodiments, the predetermined condition includes updating the recommendation count value in the event that the media information selected from the original media library is of a predetermined type after the media information selected from the original media library is recommended to the user. That is, in the second updating mode, if the media information selected from the original media library is of a predetermined type, the current recommendation count value is increased by 1 so as to update the recommendation count value. Conversely, if the media information selected from the original media library is not of the predetermined type, the recommendation count value is not updated.
It should be noted that, since it is not the inventive point of the present disclosure to recommend media information from the original media library by using the original recommendation policy, the description is not made here.
If the media information recommendation mode is the second recommendation mode, the second processing module 32 selects a target record associated with the user from the repeated recommendation list, pushes media information associated with the media information identifier in the target record to the user, and resets the recommendation count value.
In some embodiments, if multiple target records associated with the user are included in the repeated recommendation list, one target record may be randomly selected.
In other embodiments, if the recommendation count value is updated only for the predetermined type of media information in the second updating manner, in a case that the repeated recommendation list includes a plurality of target records associated with the user, the type of the media information associated with the media information identifier in the selected target record is different from the predetermined type.
In some embodiments, the remaining number of recommendations in the target record is updated after the media information associated with the media information identification is pushed to the user. And if the residual recommendation times in the target record are 0, deleting the target record.
Fig. 4 is a schematic structural diagram of an information recommendation device according to another embodiment of the present disclosure. Fig. 4 is different from fig. 3 in that, in the embodiment shown in fig. 4, the information recommendation apparatus further includes a third processing module 33.
The third processing module 33 collects the user behavior of the user for the currently played media information, and determines whether the currently played media information needs to be written into the repeated recommendation list according to the user behavior.
In some embodiments, the third processing module 33 builds the historical behavior library by aggregating behaviors that are of interest to media information and behaviors that are not of interest to media information when different users are viewing the media information. For example, actions that a user is interested in media information include: focusing on media information, viewing in cycles, viewing other media information of the same anchor, viewing comments, pausing, viewing homogeneous media information, searching for media information, forwarding media information, adjusting volume, body motion, and the like. The behavior of the user not interested in the media information comprises: the method comprises the steps of not seeing the video, quickly swiping, clicking a dislike button, swiping after seeing the same type of media information, not opening the media information and the like.
In some embodiments, the third processing module 33 builds a user interest behavior library for each user by replicating a historical behavior library.
In some embodiments, the third processing module 33 detects whether the user behavior includes a behavior that is not interested in the media information by using the user interest behavior library. And if the user behaviors comprise behaviors which are not interested in the media information, determining that the currently played media information does not need to be written into the repeated recommendation list. And if the user behaviors do not include the behaviors which are not interested in the media information, further detecting whether the user behaviors include the behaviors which are interested in the media information. And if the user behaviors comprise the behaviors interested in the media information, determining that the currently played media information needs to be written into a repeated recommendation list.
That is, as long as there is an action of no interest, the currently played media information is not written in the repeated recommendation list. Or, if only the behavior of interest exists, the currently played media information is written into the repeated recommendation list.
In other embodiments, the third processing module 33 obtains a score corresponding to the user behavior by using the user interest behavior library, and determines the total value according to the obtained score. And determining whether the currently played media information needs to be written into the repeated recommendation list according to the summary value.
For example, if the summary value associated with the behavior interested in the media included in the summary value is greater than the preset summary threshold, it is determined that the currently played media information needs to be written into the repeated recommendation list.
For another example, if a ratio of the summary value associated with the behavior interested in the media included in the summary value to the summary value associated with the behavior not interested in the media included in the summary value is greater than a preset ratio threshold, it is determined that the currently played media information needs to be written into the repeated recommendation list.
If it is determined that the currently played media information needs to be written into the duplicate recommendation list, the third processing module 33 queries whether a record associated with the user identifier of the user and the information identifier of the currently played media information is included in the duplicate recommendation list. If the repeated recommendation list does not include a record associated with the user identifier of the user and the information identifier of the currently played media information, the third processing module 33 creates a new record in the repeated recommendation list, where the new record includes the user identifier of the user, the information identifier of the currently played media information, and the preset remaining recommendation times.
If the repeated recommendation list includes a record associated with the user identifier of the user and the information identifier of the currently played media information, the third processing module 33 updates the remaining recommendation times to the preset remaining recommendation times in the record associated with the user identifier of the user and the information identifier of the currently played media information.
Fig. 5 is a schematic structural diagram of an information recommendation device according to yet another embodiment of the present disclosure. As shown in fig. 5, the information recommendation apparatus includes a memory 51 and a processor 52.
The memory 51 is used for storing instructions, the processor 52 is coupled to the memory 51, and the processor 52 is configured to execute the method according to any one of the embodiments in fig. 1 or fig. 2 based on the instructions stored in the memory.
As shown in fig. 5, the information recommendation apparatus further includes a communication interface 53 for information interaction with other devices. Meanwhile, the information recommendation device further comprises a bus 54, and the processor 52, the communication interface 53 and the memory 51 are communicated with each other through the bus 54.
The memory 51 may comprise a high-speed RAM memory, and may also include a non-volatile memory (non-volatile memory), such as at least one disk memory. The memory 51 may also be a memory array. The storage 51 may also be partitioned and the blocks may be combined into virtual volumes according to certain rules.
Further, the processor 52 may be a central processing unit CPU, or may be an application specific integrated circuit ASIC, or one or more integrated circuits configured to implement embodiments of the present disclosure.
The present disclosure also relates to a computer-readable storage medium, wherein the computer-readable storage medium stores computer instructions, and the instructions, when executed by a processor, implement a method according to any one of the embodiments shown in fig. 1 or fig. 2.
In some embodiments, the functional unit modules described above can be implemented as a general purpose Processor, a Programmable Logic Controller (PLC), a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), a Field Programmable Gate Array (FPGA) or other Programmable Logic device, discrete Gate or transistor Logic, discrete hardware components, or any suitable combination thereof for performing the functions described in this disclosure.
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 description of the present disclosure has been presented for purposes of illustration and description, and is not intended to be exhaustive or limited to the disclosure in the form disclosed. Many modifications and variations will be apparent to practitioners skilled in this art. The embodiment was chosen and described in order to best explain the principles of the disclosure and the practical application, and to enable others of ordinary skill in the art to understand the disclosure for various embodiments with various modifications as are suited to the particular use contemplated.

Claims (17)

1. An information recommendation method, comprising:
responding to information recommendation triggering operation performed by a user, and determining a media information recommendation mode according to a recommendation count value;
if the media information recommendation mode is a first recommendation mode, selecting a piece of media information from an original media library by using an original recommendation strategy and recommending the media information to the user, and adding 1 to the recommendation count value according to a preset condition so as to update the recommendation count value;
if the media information recommendation mode is a second recommendation mode, selecting a target record associated with the user from a repeated recommendation list;
and pushing the media information associated with the media information identifier in the target record to the user, and resetting the recommendation count value.
2. The method of claim 1, further comprising:
after the media information associated with the media information identifier is pushed to the user, updating the residual recommendation times in the target record;
and if the residual recommendation times in the target record are 0, deleting the target record.
3. The method of claim 1, wherein determining a media information recommendation based on the recommendation count value comprises:
judging whether the recommended count value is smaller than a preset count threshold value or not;
if the recommendation count value is smaller than the preset count threshold value, determining that the media information recommendation mode is the first recommendation mode;
and if the recommendation count value is equal to the preset count threshold value, determining that the media information recommendation mode is the second recommendation mode.
4. The method of claim 3, wherein,
the preset condition includes that after the media information selected from the original media library is recommended to the user, the recommendation count value is added with 1 to update the recommendation count value.
5. The method of claim 4, wherein,
the preset condition further includes that after the media information selected from the original media library is recommended to the user, under the condition that the media information selected from the original media library is of a preset type, the recommendation count value is added with 1 to update the recommendation count value.
6. The method of claim 5, wherein,
the type of media information associated with the media information identification is different from the predetermined type.
7. The method of claim 1, wherein determining a media information recommendation based on the recommendation count value comprises:
configuring a first probability for the first recommendation mode and a second probability for the second recommendation mode, wherein the sum of the first probability and the second probability is 1, and the second probability and the recommendation count value have positive correlation;
selecting the first recommendation mode and the second recommendation mode according to the first probability and the second probability;
and determining the media information recommendation mode according to the selection result.
8. The method of any of claims 1-7, further comprising:
collecting user behaviors of the user aiming at the currently played media information;
determining whether the currently played media information needs to be written into a repeated recommendation list according to the user behavior;
if the currently played media information is determined to be required to be written into a repeated recommendation list, inquiring whether a record associated with the user identification of the user and the information identification of the currently played media information is included in the repeated recommendation list;
and if the repeated recommendation list does not comprise a record associated with the user identifier of the user and the information identifier of the currently played media information, creating a new record in the repeated recommendation list, wherein the new record comprises the user identifier of the user, the information identifier of the currently played media information and the preset residual recommendation times.
9. The method of claim 8, further comprising:
and if the repeated recommendation list comprises records associated with the user identification of the user and the information identification of the currently played media information, updating the residual recommendation times to preset residual recommendation times in the records associated with the user identification of the user and the information identification of the currently played media information.
10. The method of claim 8, wherein determining whether currently playing media information needs to be written into a duplicate recommendation list based on the user behavior comprises:
detecting whether the user behaviors comprise behaviors which are not interested in media information;
if the user behavior comprises a behavior which is not interested in the media information, determining that the currently played media information does not need to be written into a repeated recommendation list;
if the user behaviors do not include behaviors which are not interested in the media information, further detecting whether the user behaviors include behaviors which are interested in the media information;
and if the user behaviors include the behaviors interested in the media information, determining that the currently played media information needs to be written into a repeated recommendation list.
11. The method of claim 8, wherein determining whether currently playing media information needs to be written into a duplicate recommendation list based on the user behavior comprises:
obtaining a score corresponding to the user behavior by using a user interest behavior library, and determining a summary value according to the obtained score, wherein the user interest behavior library is constructed by using a historical behavior library, and the user interest behavior library comprises behaviors which are interested in media and corresponding scores, and behaviors which are not interested in the media and corresponding scores;
and determining whether the currently played media information needs to be written into a repeated recommendation list according to the summary value.
12. The method of claim 11, wherein determining whether currently playing media information needs to be written into a duplicate recommendation list based on the summary value comprises:
and if the summary value associated with the behavior interested in the media, which is included in the summary value, is greater than a preset summary threshold value, determining that the currently played media information needs to be written into a repeated recommendation list.
13. The method of claim 11, wherein determining whether currently playing media information needs to be written into a duplicate recommendation list based on the summary value comprises:
and if the ratio of the summary value associated with the behavior interested in the media in the summary value to the summary value associated with the behavior not interested in the media in the summary value is greater than a preset proportion threshold, determining that the currently played media information needs to be written into a repeated recommendation list.
14. The method of claim 11, further comprising:
after media information associated with the media information identification is pushed to the user, collecting the watching behavior of the user;
and updating the scores of the corresponding behaviors in the user interest behavior library of the user and the scores of the corresponding behaviors in the historical behavior library according to the watching behaviors of the user.
15. An information recommendation apparatus comprising:
the first processing module is configured to respond to information recommendation triggering operation performed by a user and determine a media information recommendation mode according to a recommendation count value;
the second processing module is configured to select a piece of media information from an original media library by using an original recommendation strategy and recommend the piece of media information to the user if the media information recommendation mode is the first recommendation mode, and add 1 to the recommendation count value according to a preset condition so as to update the recommendation count value; and if the media information recommendation mode is a second recommendation mode, selecting a target record associated with the user from a repeated recommendation list, pushing media information associated with the media information identifier in the target record to the user, and resetting the recommendation count value.
16. An information recommendation apparatus comprising:
a memory configured to store instructions;
a processor coupled to the memory, the processor configured to perform implementing the method of any of claims 1-14 based on instructions stored by the memory.
17. A computer-readable storage medium, wherein the computer-readable storage medium stores computer instructions which, when executed by a processor, implement the method of any one of claims 1-14.
CN202110475167.3A 2021-04-29 2021-04-29 Information recommendation method and device Pending CN113139086A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202110475167.3A CN113139086A (en) 2021-04-29 2021-04-29 Information recommendation method and device

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202110475167.3A CN113139086A (en) 2021-04-29 2021-04-29 Information recommendation method and device

Publications (1)

Publication Number Publication Date
CN113139086A true CN113139086A (en) 2021-07-20

Family

ID=76816428

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202110475167.3A Pending CN113139086A (en) 2021-04-29 2021-04-29 Information recommendation method and device

Country Status (1)

Country Link
CN (1) CN113139086A (en)

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20140280140A1 (en) * 2013-03-14 2014-09-18 The Directv Group, Inc. Method and system for displaying recommended content such as movies associated with a cluster
CN106326431A (en) * 2016-08-25 2017-01-11 乐视控股（北京）有限公司 Information recommendation method and device
CN109034935A (en) * 2018-06-06 2018-12-18 平安科技（深圳）有限公司 Products Show method, apparatus, computer equipment and storage medium

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20140280140A1 (en) * 2013-03-14 2014-09-18 The Directv Group, Inc. Method and system for displaying recommended content such as movies associated with a cluster
CN106326431A (en) * 2016-08-25 2017-01-11 乐视控股（北京）有限公司 Information recommendation method and device
CN109034935A (en) * 2018-06-06 2018-12-18 平安科技（深圳）有限公司 Products Show method, apparatus, computer equipment and storage medium
WO2019232822A1 (en) * 2018-06-06 2019-12-12 平安科技（深圳）有限公司 Product recommendation method and apparatus, computer device, and storage medium

Non-Patent Citations (3)

* Cited by examiner, † Cited by third party
Title
DEZHI KONG; YONGQUAN FAN; YAJUN DU; SONGJIANG HU; YANG LIU; QING LI: "Personalized recommendation algorithm based on the chance discovery in social network services", IEEE, 14 April 2019 (2019-04-14) *
房小可;严承希;: "融合情境关系的社会化媒体用户兴趣推荐", 图书情报工作, no. 21, 5 November 2017 (2017-11-05) *
罗叶贝;黄昭婷;张延华;: "混合式流动放映电影推荐算法", 现代电影技术, no. 04, 11 April 2017 (2017-04-11) *

Similar Documents

Publication Publication Date Title
CN106326391B (en) Multimedia resource recommendation method and device
CN106649647B (en) Search result ordering method and device based on artificial intelligence
CN112364202B (en) Video recommendation method and device and electronic equipment
US8533202B2 (en) Entropy-based mixing and personalization
EP3036657A2 (en) Systems, methods, and computer readable medium for generating playlists
CN111432226B (en) Live broadcast recommendation method and device, server, terminal and storage medium
JP6005716B2 (en) Information processing system, server, program, and information processing method
CN103279513A (en) Method for generating content label and method and device for providing multi-media content information
CN107729578B (en) Music recommendation method and device
CN106599047B (en) Information pushing method and device
US9147435B2 (en) Weighted playlist
US10394838B2 (en) App store searching
CN106156351B (en) Multimedia resource recommendation information generation method and device
CN107748801B (en) News recommendation method and device, terminal equipment and computer readable storage medium
CN105898426A (en) Multimedia content processing method and device and server
CN113139086A (en) Information recommendation method and device
JP7350364B2 (en) Long-tail keyword identification method, keyword search method and computer equipment performed by computer equipment
US9406083B2 (en) Ranked navigation element
CN110704139B (en) Icon classification method and device
CN111460181A (en) Multimedia resource recommendation method, electronic device and storage medium
CN113139087A (en) Information recommendation method and device
CN113449147A (en) Video recommendation method and device based on theme
CN110825939A (en) Method and device for generating and sorting scores of posts, electronic equipment and storage medium
CN107404657B (en) Advertisement recommendation method and device
CN105260425A (en) Cloud disk based file display method and apparatus

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