WO2019134285A1 - 一种直播间推荐方法、电子设备及可读存储介质 - Google Patents

一种直播间推荐方法、电子设备及可读存储介质 Download PDF

Info

Publication number
WO2019134285A1
WO2019134285A1 PCT/CN2018/082166 CN2018082166W WO2019134285A1 WO 2019134285 A1 WO2019134285 A1 WO 2019134285A1 CN 2018082166 W CN2018082166 W CN 2018082166W WO 2019134285 A1 WO2019134285 A1 WO 2019134285A1
Authority
WO
WIPO (PCT)
Prior art keywords
live broadcast
live
user
broadcast set
probability
Prior art date
Application number
PCT/CN2018/082166
Other languages
English (en)
French (fr)
Inventor
王璐
陈少杰
张文明
Original Assignee
武汉斗鱼网络科技有限公司
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 武汉斗鱼网络科技有限公司 filed Critical 武汉斗鱼网络科技有限公司
Publication of WO2019134285A1 publication Critical patent/WO2019134285A1/zh

Links

Images

Classifications

    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04NPICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N21/00Selective content distribution, e.g. interactive television or video on demand [VOD]
    • H04N21/20Servers specifically adapted for the distribution of content, e.g. VOD servers; Operations thereof
    • H04N21/21Server components or server architectures
    • H04N21/218Source of audio or video content, e.g. local disk arrays
    • H04N21/2187Live feed
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/90Details of database functions independent of the retrieved data types
    • G06F16/95Retrieval from the web
    • G06F16/953Querying, e.g. by the use of web search engines
    • G06F16/9535Search customisation based on user profiles and personalisation
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04NPICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N21/00Selective content distribution, e.g. interactive television or video on demand [VOD]
    • H04N21/40Client devices specifically adapted for the reception of or interaction with content, e.g. set-top-box [STB]; Operations thereof
    • H04N21/43Processing of content or additional data, e.g. demultiplexing additional data from a digital video stream; Elementary client operations, e.g. monitoring of home network or synchronising decoder's clock; Client middleware
    • H04N21/442Monitoring of processes or resources, e.g. detecting the failure of a recording device, monitoring the downstream bandwidth, the number of times a movie has been viewed, the storage space available from the internal hard disk
    • H04N21/44204Monitoring of content usage, e.g. the number of times a movie has been viewed, copied or the amount which has been watched
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04NPICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N21/00Selective content distribution, e.g. interactive television or video on demand [VOD]
    • H04N21/40Client devices specifically adapted for the reception of or interaction with content, e.g. set-top-box [STB]; Operations thereof
    • H04N21/45Management operations performed by the client for facilitating the reception of or the interaction with the content or administrating data related to the end-user or to the client device itself, e.g. learning user preferences for recommending movies, resolving scheduling conflicts
    • H04N21/466Learning process for intelligent management, e.g. learning user preferences for recommending movies
    • H04N21/4668Learning process for intelligent management, e.g. learning user preferences for recommending movies for recommending content, e.g. movies

Definitions

  • the present invention relates to the field of electronic technologies, and in particular, to a live broadcast recommendation method, an electronic device, and a readable storage medium.
  • the user's change in interest in the live broadcast is constantly changing over time and the user's own behavior.
  • a more common strategy is to recommend to the user that they have seen a similar live broadcast room in the live broadcast room.
  • the user's viewing intentions for similar live broadcasts will change.
  • Directly recommending these similar live broadcasts may have side effects. For example, a user sees a live broadcast of a live game game for a period of time. At this time, the user may not want to watch the game again if he withdraws from his real intention. If another live broadcast room of the game match is recommended to the user, the user will broadcast the recommendation. The viewing intention between the two is actually weak, and the effectiveness of the live broadcast recommended to the user is poor.
  • An embodiment of the present invention provides a live broadcast recommendation method, an electronic device, and a readable storage medium, which are used to provide a live function recommendation method based on a memory function, and accurately measure a user's similarity after watching a live broadcast room through a memory function. Changes in interest in the live broadcast, combined with changes in user interest, provide users with an effective and reliable recommended live room.
  • the present invention provides a method for recommending a live broadcast, including:
  • the live application When the live application is enabled, determining a first live broadcast set viewed by the user in the first preset time range, where the first live broadcast set corresponds to multiple second live broadcast sets, and the second live broadcast room
  • the collection is a collection of live rooms similar to the first live broadcast set
  • a re-view probability of each second live broadcast set in the plurality of second live broadcast sets wherein the memory function is used to indicate that the viewer views the first live broadcast set at the current time
  • the probability of viewing is a probability that the first live broadcast set is viewed in the first preset time range and the second live broadcast set is viewed in a second preset time range;
  • the calculating based on the memory function, obtaining a revisiting probability of each second live broadcast set in the plurality of second live broadcast sets, including:
  • is an attenuation coefficient
  • T1 is a time point set in which the user u views the first live broadcast set s in the second time range
  • the current time value of the whole network user to the first live broadcast set is
  • the live broadcast room that meets the preset condition in the second live broadcast set with the highest probability of revisiting is pushed to the user, including:
  • the similarity between the second live broadcast set and the first live broadcast that is the highest probability of revisiting is greater than a first preset threshold.
  • the second live room is pushed to the user.
  • the live broadcast room that meets the preset condition in the second live broadcast set with the highest probability of revisiting is pushed to the user, including:
  • the similarity between the second live broadcast set with the highest probability of revisiting and the live broadcast of the first live broadcast set is greater than the first
  • the second live broadcast of the preset threshold is pushed to the user.
  • the viewing duration of each live broadcast in the first live broadcast set is greater than a preset duration.
  • an embodiment of the present invention provides an electronic device, where the electronic device includes:
  • a first determining unit configured to determine, in a case where the live application is enabled, a first live broadcast set viewed by the user in the first preset time range, where the first live broadcast set corresponds to multiple second live broadcast sets
  • the second live broadcast set is a live broadcast set similar to the first live broadcast set
  • a calculating unit configured to calculate, according to a memory function, a re-view probability of each second live broadcast set in the plurality of second live broadcast sets, where the memory function is used to indicate that the first live broadcast is performed by the viewer at the current moment a viewing probability of the set, the probability of viewing is a probability of viewing the first live broadcast set in the first preset time range and viewing the second live broadcast set in a second preset time range;
  • a second determining unit configured to determine a second live broadcast set having the highest probability of seeing again in the plurality of second live broadcast sets
  • a pushing unit configured to push, to the user, a live room that meets a preset condition in the second live broadcast set with the highest probability of revisiting.
  • the computing unit is configured to:
  • is an attenuation coefficient
  • T1 is a time point set in which the user u views the first live broadcast set s in the second time range
  • the current time value of the whole network user to the first live broadcast set is
  • the pushing unit is configured to: after the user views the first live broadcast room in the first live broadcast set, the second live broadcast set with the highest probability of revisiting and the first The second live broadcast room with the similarity between the live broadcasts being greater than the first preset threshold is pushed to the user.
  • the pushing unit is configured to: after the user views any one of the first live broadcast sets, the first live broadcast set with the highest probability of revisiting and the first The second live broadcast room in which the similarity between the live broadcasts in the live broadcast set is greater than the second preset threshold is pushed to the user.
  • the viewing duration of each live broadcast in the first live broadcast set is greater than a preset duration.
  • an embodiment of the present invention provides an electronic device, where the electronic device includes a processor, and when the processor is configured to execute a computer program stored in a memory, the live broadcast recommendation as described in the foregoing first embodiment is implemented. The steps of the method.
  • an embodiment of the present invention provides a readable storage medium, where a computer program is stored, and when the computer program is executed by a processor, the method for recommending a live broadcast according to the foregoing first embodiment is implemented. step.
  • the memory function is used to accurately measure the revisiting probability of the user watching the similar second live broadcast set after watching the first live broadcast set. After calculating the revisiting probability of each similar second live broadcast set, the selection is performed. The second live broadcast set with the highest probability is displayed, and the live broadcast room that meets the preset condition in the second live broadcast set with the highest probability is pushed to the user. In this way, the memory function is used to accurately measure the interest changes of the user after watching the live broadcast room, and fully consider the user's interest change, and provide the user with an effective and reliable recommended live broadcast room.
  • FIG. 1 is a flowchart of a method for recommending a live broadcast in a first embodiment of the present invention
  • FIG. 2 is a schematic diagram of an electronic device in a second embodiment of the present invention.
  • FIG. 3 is a schematic diagram of an electronic device in a third embodiment of the present invention.
  • An embodiment of the present invention provides a live broadcast recommendation method, an electronic device, and a readable storage medium, which are used to provide a live function recommendation method based on a memory function, and accurately measure a user's similarity after watching a live broadcast room through a memory function. Changes in interest in the live broadcast, combined with changes in user interest, provide users with an effective and reliable recommended live room.
  • the recommendation method includes: determining, in a case where the live application is enabled, a first live broadcast set viewed by the user in the first preset time range, where the first live broadcast set corresponds to multiple second live broadcast sets, The second live broadcast set is a live broadcast set similar to the first live broadcast set; and based on the memory function, the revisit probability of each second live broadcast set in the plurality of second live broadcast sets is obtained.
  • the memory function is used to indicate the viewer's viewing memory of the first live broadcast set at the current time, and the replay probability is that the first live broadcast set is viewed in the first preset time range.
  • Determining a probability of the second live broadcast set in a preset time range determining a second live broadcast set having the highest probability of seeing again in the plurality of second live broadcast sets; and the second live broadcast room having the highest probability of seeing again A live room in the set that meets the preset condition is pushed to the user.
  • a first embodiment of the present invention provides a method for recommending a live broadcast, including the following steps:
  • S101 Determine, in a case that the live application is enabled, the first live broadcast set viewed by the user in the first preset time range, where the first live broadcast set corresponds to multiple second live broadcast sets, and the second The live rooms are collected as a set of live rooms similar to the first live set;
  • S102 Calculate, according to a memory function, a re-view probability of each second live broadcast set in the plurality of second live broadcast sets, where the memory function is used to indicate that the viewer collects the first live broadcast at the current time.
  • the probability of revisiting is a probability of viewing the first live broadcast set in the first preset time range and viewing the second live broadcast set in a second preset time range;
  • S104 Push the live broadcast room that meets the preset condition in the second live broadcast set with the highest probability of revisiting to the user.
  • the recommended method in this embodiment may be applied to the server end of the live application.
  • the server determines that the client corresponding to a certain user starts the live application, the server stores the history of each user, and then the server is In step S101, by calling the history record with the user, the first live broadcast set S that the user viewed in the first preset time range is determined.
  • the recommended method in this embodiment may be applied to a client that installs a live application, and the client may be an electronic device such as a mobile phone, a tablet computer, a desktop computer, or other electronic devices. limit.
  • the server may request the history record corresponding to the user of the client, and then determine the first live broadcast set s that the user viewed in the first preset time range in step S101.
  • the first preset time range may be set according to actual needs, for example, may be set to the first 5 days of the closest 10 days from the current time, for example, there is a closest 10 days from the current time. From the farthest to the current time, including the first day to the tenth day, the first preset time range is from the first day to the fifth day.
  • the first live broadcast set s that the user viewed in the first day to the fifth day is counted.
  • the first live broadcast set s corresponds to a plurality of similar second live broadcast sets s*, and each second live broadcast set s* is different from each other.
  • step S102 by using a memory function, the user changes the interest of the similar second live broadcast set s* after viewing the first live broadcast set, that is, based on the memory function calculation.
  • the first live broadcast set s corresponds to three similar second live broadcast sets, namely s1*, s2*, and s3*.
  • the revisit probability Ps1* of the second live broadcast set s1* may be 0.6 based on the memory function, the revisit probability Ps2* of the second live broadcast set s2* is 0.8, and the revisit probability Ps3* of the second live broadcast set s3* Is 0.2.
  • the second time range is after the first time range, for example, there is the closest 10 days from the current time, and the first preset time range is the first time from the current time from far to near, including the first day to the tenth day. On the fifth day, the second time range is from day 6 to day 10.
  • the first time range and the second time range may be set according to actual needs, and the present application does not limit the present application.
  • the second live broadcast set with the highest probability of revisiting is determined, such as the second live broadcast set s1*, the second live broadcast set s2*, and the second live broadcast set s3* in the above example.
  • the highest probability is the second live set s2*.
  • the step S104 can be used to view the second live broadcast set with the highest probability.
  • the live broadcast room that meets the preset conditions is selected and pushed to the user, and since the change of the user interest is considered, the live broadcast room recommended for the user can be made more effective.
  • the history record of the network user is measured, which is a default memory function related only to the first live broadcast set s, and reflects the potential viewing possibilities of the first live broadcast set s.
  • N(s) is the total number of people watching the first live broadcast set s in the first preset time range
  • N(s * , T) is the first time viewed in the first preset time range
  • the number of the second live broadcast set s* is viewed in the second preset time range T after the viewing
  • T is the length of the memory time, which can be set according to actual needs.
  • Mu(u, t, s) is the viewing memory of the user u at the current time t for the first live set s, and reflects the potential viewing possibilities of the individual for the first live set s.
  • is the attenuation coefficient and is a constant greater than zero.
  • T1 is a time point set in which the user u views the first live broadcast set s in the second time range T. Since the user's memory residue between the live broadcasts decays with time, the viewing memory uses an exponential distribution. To make an estimate.
  • the memory value of the current live time set s can be obtained for a single user, and thus can be obtained therefrom.
  • the memory value of the entire network user for the first live broadcast set s That is, the sum of the memory values of all users in the first time range for the first live set s is counted. In this way, the memory value of the entire network user for the first live broadcast set s corresponding to the current time set of each second live broadcast set can be obtained.
  • the revisiting probability corresponding to the second live broadcast set is the number of people watching the second live broadcast set at the current time divided by the memory value of the first live broadcast set by the entire network user at the current time and in the second time range The difference in the number of people who have watched the second live set.
  • the formula for the probability of revisiting the second live broadcast set can be expressed as:
  • the calculation of the revisit probability of the second live set is explained in detail below with a complete example.
  • the first preset time range is from the first day to the fifth day
  • the second time range is from the sixth day to Day 10.
  • the user views the first live broadcast set s, and the second live broadcast set s4* similar to the first live broadcast set s, through the viewing record from the sixth day to the tenth day, the first record can be obtained.
  • the total number of people watching the first live room collection s from the first day to the fifth day is 10000, and the first live room collection s is watched from the first day to the fifth day, and then viewed from the sixth day to the tenth day.
  • the viewing duration is less than the preset duration (eg, 1 minute, 2 minutes, etc.)
  • the viewing duration is less than the preset duration (eg, 1 minute, 2 minutes, etc.)
  • the viewing duration is less than the preset duration (eg, 1 minute, 2 minutes, etc.).
  • the preset duration eg, 1 minute, 2 minutes, etc.
  • step S104 the following two inter-live push modes may be adopted:
  • the first type after the user views the first live broadcast room in the first live broadcast set, the similarity between the second live broadcast set and the first live broadcast that is the highest probability of revisiting is greater than the first A second live broadcast of a preset threshold is pushed to the user.
  • the second live broadcast set with the highest probability of revisiting is determined according to the foregoing manner, and then the most probable
  • the second live broadcast set may include a plurality of second live broadcast rooms B similar to the first live broadcast room A, and the similarity between the plurality of second live broadcast rooms B and the first live broadcast room A is greater than a first preset threshold.
  • the live broadcast is pushed to the user.
  • the first preset threshold can be set to a value of 0.5, 0.6, etc., of course, other values can be set as needed, and the present application does not limit the application.
  • the second type after the user views any one of the first live broadcast sets, the second live broadcast set with the highest probability of revisiting and the live broadcast of the first live broadcast set
  • the second live broadcast room with the similarity greater than the second preset threshold is pushed to the user.
  • the second live broadcast set with the highest probability of revisiting is determined according to the foregoing manner, if the first live broadcast set includes Live room A1, live room A2, live room A3.
  • the second live broadcast set with the highest probability is the live room B1, the live room B2, the live room B3, the live room B4, and the live room B5.
  • the live room B1 and the live room B2 are similar to the live room A1.
  • the similarity between the live broadcast room B1 and the live broadcast room A1 is 0.4, and the similarity between the live broadcast room B2 and the live broadcast room A1 is 0.7.
  • the live broadcast room B3 and the live broadcast room B4 are similar to the live broadcast room A2.
  • the similarity between the live broadcast room B3 and the live broadcast room A2 is 0.65, and the similarity between the live broadcast room B4 and the live broadcast room A2 is 0.5.
  • the live broadcast room B5 is a live broadcast room similar to the live broadcast room A3, and the similarity between the live broadcast room B5 and the live broadcast room A3 is 0.75.
  • the live broadcast between the second live broadcast set that has the highest probability of being observed and the live broadcast of the first live broadcast set is greater than the second preset threshold, and the second preset threshold is set to 0.6. You need to recommend the live room B2, the live room B3, and the live room B5 to the user.
  • the recommended method is not limited to the above two types, and other methods may also be used, for example, all live rooms in the second live broadcast set with the highest probability of being viewed are pushed to the user. No limitation is imposed on this application.
  • a third embodiment of the present invention provides an electronic device, where the electronic device includes:
  • the first determining unit 201 is configured to determine, in the case that the live application is enabled, the first live broadcast set viewed by the user in the first preset time range, where the first live broadcast set corresponds to multiple second live broadcast rooms.
  • the collection, the second live broadcast set is a live broadcast set similar to the first live broadcast set;
  • the calculating unit 202 is configured to calculate, according to a memory function, a re-view probability of obtaining each second live broadcast set in the plurality of second live broadcast sets, where the memory function is used to indicate that the viewer is at the current moment
  • the viewing memory of the live broadcast the probability of viewing is the probability that the first live broadcast set is viewed in the first preset time range and the second live broadcast set is viewed in the second preset time range.
  • a second determining unit 203 configured to determine a second live broadcast set with the highest probability of revisiting among the plurality of second live broadcast sets
  • the pushing unit 204 is configured to push the live room that meets the preset condition in the second live broadcast set with the highest probability of revisiting to the user.
  • the electronic device in this embodiment may be a server of a live application.
  • the server determines that the client corresponding to a certain user starts the live application, the server stores the history of each user, and then the server determines by using the first determination.
  • the unit 201 invokes a history record with the user to determine the first live broadcast set S that the user has viewed in the first preset time range.
  • the electronic device in this embodiment may be a client that installs a live application, and the client may be an electronic device such as a mobile phone, a tablet computer, a desktop computer, or other electronic device, and the present application does not limit the application.
  • the server may request the history record corresponding to the user of the client, and then determine, by the first determining unit 201, the first live broadcast set viewed by the user in the first preset time range. s.
  • the first preset time range may be set according to actual needs, for example, may be set to the first 5 days of the closest 10 days from the current time, for example, there is a closest 10 days from the current time. From the farthest to the current time, including the first day to the tenth day, the first preset time range is from the first day to the fifth day.
  • the first live broadcast set s that the user viewed in the first day to the fifth day is counted.
  • the first live broadcast set s corresponds to a plurality of similar second live broadcast sets s*, and each second live broadcast set s* is different from each other.
  • the electronic device in this embodiment uses the memory function to measure the change of interest of the user to the similar second live broadcast set s* after viewing the first live broadcast set by the computing unit 202, that is, based on the memory function degree calculation. Obtaining a revisiting probability that the user views the first live broadcast set in the first preset time range and the second live broadcast set s* in the second preset time range.
  • the first live broadcast set s corresponds to three similar second live broadcast sets, namely s1*, s2*, and s3*.
  • the revisit probability Ps1* of the second live broadcast set s1* may be 0.6 based on the memory function, the revisit probability Ps2* of the second live broadcast set s2* is 0.8, and the revisit probability Ps3* of the second live broadcast set s3* Is 0.2.
  • the second time range is after the first time range, for example, there is the closest 10 days from the current time, and the first preset time range is the first time from the current time from far to near, including the first day to the tenth day. On the fifth day, the second time range is from day 6 to day 10.
  • the first time range and the second time range may be set according to actual needs, and the present application does not limit the present application.
  • the second determining unit 203 determines the second live broadcast set with the highest probability of revisiting, such as the second live broadcast set s1*, the second live broadcast set s2*, and the second live broadcast set s3* in the above example.
  • the highest probability of seeing is the second live broadcast set s2*.
  • the second live broadcast set with the highest probability can be seen by the push unit 204.
  • the live broadcast room that meets the preset conditions is selected and pushed to the user, and since the change of the user interest is considered, the live broadcast room recommended for the user can be made more effective.
  • the historical watch record is a default memory function related only to the first live set s, which reflects the potential viewing possibilities of the public for the first live set s.
  • N(s) is the total number of people watching the first live broadcast set s in the first preset time range
  • N(s * , T) is the first time viewed in the first preset time range
  • the number of the second live broadcast set s* is viewed in the second preset time range T after the viewing
  • T is the length of the memory time, which can be set according to actual needs.
  • Mu(u, t, s) is the viewing memory of the user u at the current time t for the first live set s, and reflects the potential viewing possibilities of the individual for the first live set s.
  • is the attenuation coefficient and is a constant greater than zero.
  • T1 is a time point set in which the user u views the first live broadcast set s in the second time range T. Since the user's memory residue between the live broadcasts decays with time, the viewing memory uses an exponential distribution. To make an estimate.
  • the memory value of the current live time set s can be obtained for a single user, and thus can be obtained therefrom.
  • the memory value of the entire network user for the first live broadcast set s That is, the sum of the memory values of all users in the first time range for the first live set s is counted. In this way, the memory value of the entire network user for the first live broadcast set s corresponding to the current time set of each second live broadcast set can be obtained.
  • the revisiting probability corresponding to the second live broadcast set is the number of people watching the second live broadcast set at the current time divided by the memory value of the first live broadcast set by the entire network user at the current time and in the second time range The difference in the number of people who have watched the second live set.
  • the formula for the probability of looking at the second live broadcast collection can be expressed as:
  • the calculation of the revisit probability of the second live set is explained in detail below with a complete example.
  • the first preset time range is from the first day to the fifth day
  • the second time range is from the sixth day to Day 10.
  • the user views the first live broadcast set s, and the second live broadcast set s4* similar to the first live broadcast set s, through the viewing record from the sixth day to the tenth day, the first record can be obtained.
  • the total number of people watching the first live room collection s from the first day to the fifth day is 10000, and the first live room collection s is watched from the first day to the fifth day, and then viewed from the sixth day to the tenth day.
  • the viewing duration is less than the preset duration (eg, 1 minute, 2 minutes, etc.)
  • the viewing duration is less than the preset duration (eg, 1 minute, 2 minutes, etc.)
  • the viewing duration is less than the preset duration (eg, 1 minute, 2 minutes, etc.).
  • the preset duration eg, 1 minute, 2 minutes, etc.
  • the pushing unit 204 can perform the inter-live push in the following two ways:
  • the first type after the user views the first live broadcast room in the first live broadcast set, the similarity between the second live broadcast set and the first live broadcast that is the highest probability of revisiting is greater than the first A second live broadcast of a preset threshold is pushed to the user.
  • the second live broadcast set with the highest probability of revisiting is determined according to the foregoing manner, and then the most probable
  • the second live broadcast set may include a plurality of second live broadcast rooms B similar to the first live broadcast room A, and the similarity between the plurality of second live broadcast rooms B and the first live broadcast room A is greater than a first preset threshold.
  • the live broadcast is pushed to the user.
  • the first preset threshold can be set to a value of 0.5, 0.6, etc., of course, other values can be set as needed, and the present application does not limit the application.
  • the second type after the user views any one of the first live broadcast sets, the second live broadcast set with the highest probability of revisiting and the live broadcast of the first live broadcast set
  • the second live broadcast room with the similarity greater than the second preset threshold is pushed to the user.
  • the second live broadcast set with the highest probability of revisiting is determined according to the foregoing manner, if the first live broadcast set includes Live room A1, live room A2, live room A3.
  • the second live broadcast set with the highest probability is the live room B1, the live room B2, the live room B3, the live room B4, and the live room B5.
  • the live room B1 and the live room B2 are similar to the live room A1.
  • the similarity between the live broadcast room B1 and the live broadcast room A1 is 0.4, and the similarity between the live broadcast room B2 and the live broadcast room A1 is 0.7.
  • the live broadcast room B3 and the live broadcast room B4 are similar to the live broadcast room A2.
  • the similarity between the live broadcast room B3 and the live broadcast room A2 is 0.65, and the similarity between the live broadcast room B4 and the live broadcast room A2 is 0.5.
  • the live broadcast room B5 is a live broadcast room similar to the live broadcast room A3, and the similarity between the live broadcast room B5 and the live broadcast room A3 is 0.75.
  • the live broadcast between the second live broadcast set that has the highest probability of being observed and the live broadcast of the first live broadcast set is greater than the second preset threshold, and the second preset threshold is set to 0.6. You need to recommend the live room B2, the live room B3, and the live room B5 to the user.
  • the recommended method is not limited to the above two types, and other methods may also be used, for example, all live rooms in the second live broadcast set with the highest probability of being viewed are pushed to the user. No limitation is imposed on this application.
  • a third embodiment of the present invention provides an electronic device, where the electronic device includes: a processor 301, a memory 302, and is stored in the memory and operable on the processor.
  • a computer program for example, a program corresponding to the recommendation method in the live broadcast in the first embodiment.
  • the steps in the path detection in the first embodiment described above are implemented when the processor executes the computer program.
  • the processor implements the functions of the modules/units in the electronic device of the second embodiment described above when the computer program is executed.
  • the computer program can be partitioned into one or more modules/units that are stored in the memory and executed by the processor to perform the present invention.
  • the one or more modules/units may be a series of computer program instruction segments capable of performing a particular function, the instruction segments being used to describe the execution of the computer program in the computer device.
  • the computer program may be divided into a first determining unit, a calculating unit, a second determining unit, and a pushing unit, and the specific functions of each unit are as follows:
  • a first determining unit configured to determine, in a case where the live application is enabled, a first live broadcast set viewed by the user in the first preset time range, where the first live broadcast set corresponds to multiple second live broadcast sets
  • the second live broadcast set is a live broadcast set similar to the first live broadcast set
  • a calculating unit configured to calculate, according to a memory function, a re-view probability of each second live broadcast set in the plurality of second live broadcast sets, where the memory function is used to indicate that the first live broadcast is performed by the viewer at the current moment a viewing probability of the set, the probability of viewing is a probability of viewing the first live broadcast set in the first preset time range and viewing the second live broadcast set in a second preset time range;
  • a second determining unit configured to determine a second live broadcast set having the highest probability of seeing again in the plurality of second live broadcast sets
  • a pushing unit configured to push, to the user, a live room that meets a preset condition in the second live broadcast set with the highest probability of revisiting.
  • the electronic device can include, but is not limited to, a processor, a memory. It will be understood by those skilled in the art that the schematic diagram 3 is merely an example of a computer device and does not constitute a limitation on an electronic device, and may include more or less components than those illustrated, or may combine certain components or different components.
  • the electronic device may further include an input and output device, a network access device, a bus, and the like.
  • the processor 301 may be a central processing unit (CPU), or may be other general-purpose processors, a digital signal processor (DSP), an application specific integrated circuit (ASIC), Field-Programmable Gate Array (FPGA) or other programmable logic device, discrete gate or transistor logic device, discrete hardware components, etc.
  • the general purpose processor may be a microprocessor or the processor or any conventional processor or the like, which is a control center of the computer device, which connects various parts of the entire computer device using various interfaces and lines.
  • the memory 302 can be used to store the computer program and/or module, the processor implementing the method by running or executing a computer program and/or module stored in the memory, and invoking data stored in the memory Various functions of a computer device.
  • the memory may mainly include a storage program area and a storage data area, wherein the storage program area may store an operating system, an application required for at least one function (such as a sound playing function, an image playing function, etc.), and the like; the storage data area may be stored. Data created according to the use of the mobile phone (such as audio data, video data, etc.).
  • the memory may include a high-speed random access memory, and may also include non-volatile memory such as a hard disk, a memory, a plug-in hard disk, a smart memory card (SMC), and a Secure Digital (SD) card.
  • non-volatile memory such as a hard disk, a memory, a plug-in hard disk, a smart memory card (SMC), and a Secure Digital (SD) card.
  • Flash Card at least one disk storage device, flash memory device, or other volatile solid-state storage device.
  • the historical time range includes N days closest to the current time, and N is an integer greater than 0.
  • the processor 301 included in the electronic device further has the following functions:
  • is an attenuation coefficient
  • T1 is a time point set in which the user u views the first live broadcast set s in the second time range
  • the current time value of the whole network user to the first live broadcast set is
  • processor 301 included in the electronic device further has the following functions:
  • the similarity between the second live broadcast set and the first live broadcast that is the highest probability of revisiting is greater than a first preset threshold.
  • the second live room is pushed to the user.
  • processor 301 included in the electronic device further has the following functions:
  • the similarity between the second live broadcast set with the highest probability of revisiting and the live broadcast of the first live broadcast set is greater than the first
  • the second live broadcast of the preset threshold is pushed to the user.
  • the viewing duration of each live broadcast in the first live broadcast set is greater than a preset duration.
  • a fourth embodiment of the present invention provides a computer readable storage medium having stored thereon a computer program, and the functional unit integrated by the electronic device in the second embodiment of the present invention is implemented in the form of a software functional unit and is independent When the product is sold or used, it can be stored in a computer readable storage medium.
  • the present invention implements all or part of the flow in the live broadcast recommendation method of the first embodiment described above, and may also be completed by a computer program to instruct related hardware, and the computer program may be stored in a computer readable manner.
  • the computer program when executed by the processor, implements the steps of the various method embodiments described above.
  • the computer program comprises computer program code, which may be in the form of source code, object code form, executable file or some intermediate form.
  • the computer readable medium may include any entity or device capable of carrying the computer program code, a recording medium, a USB flash drive, a removable hard disk, a magnetic disk, an optical disk, a computer memory, a read-only memory (ROM). , random access memory (RAM, Random Access Memory), electrical carrier signals, telecommunications signals, and software distribution media.
  • RAM random access memory
  • computer readable media Does not include electrical carrier signals and telecommunication signals.
  • the memory function is used to accurately measure the revisiting probability of the user watching the similar second live broadcast set after watching the first live broadcast set. After calculating the revisiting probability of each similar second live broadcast set, the selection is performed. The second live broadcast set with the highest probability is displayed, and the live broadcast room that meets the preset condition in the second live broadcast set with the highest probability is pushed to the user. In this way, the memory function is used to accurately measure the interest changes of the user after watching the live broadcast room, and fully consider the user's interest change, and provide the user with an effective and reliable recommended live broadcast room.

Landscapes

  • Engineering & Computer Science (AREA)
  • Databases & Information Systems (AREA)
  • Multimedia (AREA)
  • Signal Processing (AREA)
  • Theoretical Computer Science (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Data Mining & Analysis (AREA)
  • Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Two-Way Televisions, Distribution Of Moving Picture Or The Like (AREA)
  • Testing, Inspecting, Measuring Of Stereoscopic Televisions And Televisions (AREA)

Abstract

本发明提供了一种直播间推荐方法、电子设备及可读存储介质,用于提供一种基于记忆函数的直播间推荐方法,通过记忆函数来准确度量用户在观看直播间后对其相似直播间的兴趣变化,为用户提供有效且可靠的推荐直播间。该推荐方法包括:在直播应用程序开启情况下,确定在第一预设时间范围内用户观看过的第一直播间集合,所述第一直播间集合对应有多个相似的第二直播间集合;基于记忆函数,计算获得所述多个第二直播间集合中每个第二直播间集合的再看概率;确定所述多个第二直播间集合中再看概率最大的第二直播间集合;将所述再看概率最大的第二直播间集合中满足预设条件的直播间推送给所述用户。

Description

一种直播间推荐方法、电子设备及可读存储介质 技术领域
本发明涉及电子技术领域,尤其涉及一种直播间推荐方法、电子设备及可读存储介质。
背景技术
用户对直播间兴趣的变化是随着时间和用户自身的行为不断变化的。在现有技术的直播间推荐方法中,一个较为常见的策略是给用户推荐其看过直播间类似的直播间。然而,这样的推荐策略并不是总是有效的,用户观看了一个直播间后对相似直播间观看意图是会发生变化的,直接推荐这些相似直播间可能会产生副作用。比如:一个用户看了一个直播游戏比赛的直播间一段时间,此时用户退出其真实意图可能不想再看游戏比赛,如果再推荐直播该游戏比赛的其他直播间给用户,用户对该推荐的直播间的观看意图实际上是比较弱的,向用户推荐的直播间的有效性差。
发明内容
本发明实施例提供了一种直播间推荐方法、电子设备及可读存储介质,用于提供一种基于记忆函数的直播间推荐方法,通过记忆函数来准确度量用户在观看直播间后对其相似直播间的兴趣变化,结合用户的兴趣变化,为用户提供有效且可靠的推荐直播间。
第一方面,本发明提供了一种直播间推荐方法,包括:
在直播应用程序开启情况下,确定在第一预设时间范围内用户观看过的第一直播间集合,所述第一直播间集合对应有多个第二直播间集合,所述第二直播间集合为与所述第一直播间集合相似的直播间集合;
基于记忆函数,计算获得所述多个第二直播间集合中每个第二直播间集合的再看概率,所述记忆函数用于表示观众在当前时刻对所述第一直播间集合的观看记忆,所述再看概率为在所述第一预设时间范围内观看过所述第一直播间集合又在第二预设时间范围内观看该第二直播间集合的概率;
确定所述多个第二直播间集合中再看概率最大的第二直播间集合;
将所述再看概率最大的第二直播间集合中满足预设条件的直播间推送给所述用户。
可选的,所述基于记忆函数,计算获得所述多个第二直播间集合中每个第二直播间集合的再看概率,包括:
在计算所述多个第二直播间集合中每个第二直播间集合的再看概率时,基于记忆函数公式,确定在当前时刻全网用户对所述第一直播间集合的记忆值,以及确定在当前时刻观看该第二直播间集合的人数,以及确定在所述第二时间范围内观看过该第二直播间集合的人数,该第二直播间集合的再看概率为所述当前时刻观看该第二直播间集合的人数除以所述当前时刻全网用户对所述第一直播间集合的记忆值与所述在所述第二时间范围内观看过该第二直播间集合的人数的差值;
其中,所述记忆函数公式为m(u,t,s)=md(s)+mu(u,t,s),表示用户u在当前时刻t对第一直播间集合s的观看记忆函数,md(s)=N(s*,T)/N(s),N(s)为在第一预设时间范围内观看第一直播间集合s的总人数,N(s *,T)为在所述第一预设时间范围内观看了所述第一直播集合s的所有用户中在观看后的第二预设时间范围T内观看该第二直播间集合s*的人数,
Figure PCTCN2018082166-appb-000001
λ为衰减系数,T1为在所述第二时间范围内用户u观看所述第一直播间集合s的时间点集合,所述当前时刻全网用户对所述第一直播间集合的记忆值为
Figure PCTCN2018082166-appb-000002
可选的,所述将所述再看概率最大的第二直播间集合中满足预设条件的直播间推送给所述用户,包括:
在所述用户观看所述第一直播间集合中的第一直播间后,将所述再看概率最大的第二直播间集合中与所述第一直播间相似度大于第一预设阈值的第二直播间推送给所述用户。
可选的,所述将所述再看概率最大的第二直播间集合中满足预设条件的直播间推送给所述用户,包括:
在所述用户观看所述第一直播间集合中的任意一个直播间后,将所述再看概率最大的第二直播间集合中与所述第一直播间集合中直播间的相似度大于第二预设阈值的第二直播间推送给所述用户。
可选的,所述第一直播间集合中的每个直播间的观看时长均大于预设时长。
第二方面,本发明实施例提供一种电子设备,所述电子设备包括:
第一确定单元,用于在直播应用程序开启情况下,确定在第一预设时间范围内用户观看过的第一直播间集合,所述第一直播间集合对应有多个第二直播间集合,所述第二直播间集合为与所述第一直播间集合相似的直播间集合;
计算单元,用于基于记忆函数,计算获得所述多个第二直播间集合中每个第二直播间集合的再看概率,所述记忆函数用于表示观众在当前时刻对所述第一直播间集合的观看记忆,所述再看概率为在所述第一预设时间范围内观看过所述第一直播间集合又在第二预设时间范围内观看该第二直播间集合的概率;
第二确定单元,用于确定所述多个第二直播间集合中再看概率最大的第二直播间集合;
推送单元,用于将所述再看概率最大的第二直播间集合中满足预设条件的直播间推送给所述用户。
可选的,所述计算单元用于:
在计算所述多个第二直播间集合中每个第二直播间集合的再看概率时,基于记忆函数公式,确定在当前时刻全网用户对所述第一直播间集合的记忆值,以及确定在当前时刻观看该第二直播间集合的人数,以及确定在所述第二时间范围内观看过该第二直播间集合的人数,该第二直播间集合的再看概率为所述当前时刻观看该第二直播间集合的人数除以所述当前时刻全网用户对所述第一直播间集合的记忆值与所述在所述第二时间范围内观看过该第二直播间集合的人数的差值;
其中,所述记忆函数公式为m(u,t,s)=md(s)+mu(u,t,s),表示用户u在当前时刻t对第一直播间集合s的观看记忆函数,md(s)=N(s*,T)/N(s),N(s)为在第一预设时间范围内观看第一直播间集合s的总人数,N(s *,T)为在所述第一预设时间范围内观看了所述第一直播集合s的所有用户中在观看后的第二预设时间范围T内观看该第二直播间集合s*的人数,
Figure PCTCN2018082166-appb-000003
λ为衰减系数,T1为在所述第二时间范围内用户u观看所述第一直播间集合s的时间点集合,所述当前时刻全网用户对所述第一直播间集合的记忆值为
Figure PCTCN2018082166-appb-000004
可选的,所述推送单元用于:在所述用户观看所述第一直播间集合中的第一直播间后,将所述再看概率最大的第二直播间集合中与所述第一直播间相似度大于第一预设阈值的第二直播间推送给所述用户。
可选的,所述推送单元用于:在所述用户观看所述第一直播间集合中的任意一个直播间后,将所述再看概率最大的第二直播间集合中与所述第一直播间集合中直播间的相似度大于第二预设阈值的第二直播间推送给所述用户。
可选的,所述第一直播间集合中的每个直播间的观看时长均大于预设时长。
第三方面,本发明实施例提供一种电子设备,所述电子设备包括处理器,所述处理器用于执行存储器中存储的计算机程序时实现如前述第一方面实施例中所述的直播间推荐方法的步骤。
第四方面,本发明实施例提供了一种可读存储介质,其上存储有计算机程序,所述计算机程序被处理器执行时实现如前述第一方面实施例中所述的直播间推荐方法的步骤。
本申请实施例中的上述一个或多个技术方案,至少具有如下一种或多种技 术效果:
在本发明实施例的技术方案中,在直播应用程序开启情况下,确定在第一预设时间范围内用户观看过的第一直播间集合,该第一直播间集合包括多个相似的第二直播间集合。进而,通过记忆函数来准确度量用户在观看第一直播间集合后再观看相似的第二直播间集合的再看概率,当计算好每个相似的第二直播间集合的再看概率后,挑选出再看概率最大的第二直播间集合,将再看概率最大的第二直播间集合中满足预设条件的直播间推送给用户。这样,通过记忆函数来准确度量用户在观看直播间后对其相似直播间的兴趣变化,充分考虑了用户的兴趣变化,为用户提供有效且可靠的推荐直播间。
附图说明
图1为本发明第一实施例中的一种直播间推荐方法的流程图;
图2为本发明第二实施例中的电子设备的示意图;
图3为本发明第三实施例中电子设备的示意图。
具体实施方式
本发明实施例提供了一种直播间推荐方法、电子设备及可读存储介质,用于提供一种基于记忆函数的直播间推荐方法,通过记忆函数来准确度量用户在观看直播间后对其相似直播间的兴趣变化,结合用户的兴趣变化,为用户提供有效且可靠的推荐直播间。该推荐方法包括:在直播应用程序开启情况下,确定在第一预设时间范围内用户观看过的第一直播间集合,所述第一直播间集合对应有多个第二直播间集合,所述第二直播间集合为与所述第一直播间集合相似的直播间集合;基于记忆函数,计算获得所述多个第二直播间集合中每个第二直播间集合的再看概率,所述记忆函数用于表示观众在当前时刻对所述第一直播间集合的观看记忆,所述再看概率为在所述第一预设时间范围内观看过所述第一直播间集合又在第二预设时间范围内观看该第二直播间集合的概率;确定所述多个第二直播间集合中再看概率最大的第二直播间集合;将所述再看概率最大的第二直播间集合中满足预设条件的直播间推送给所述用户。
下面通过附图以及具体实施例对本发明技术方案做详细的说明,应当理解本申请实施例以及实施例中的具体特征是对本申请技术方案的详细的说明,而不是对本申请技术方案的限定,在不冲突的情况下,本申请实施例以及实施例中的技术特征可以相互组合。
本文中术语“和/或”,仅仅是一种描述关联对象的关联关系,表示可以存在三种关系,例如,A和/或B,可以表示:单独存在A,同时存在A和B,单独存在B这三种情况。另外,本文中字符“/”,一般表示前后关联对象是一 种“或”的关系。
实施例
请参考图1,本发明第一实施例提供一种直播间推荐方法,包括如下步骤:
S101:在直播应用程序开启情况下,确定在第一预设时间范围内用户观看过的第一直播间集合,所述第一直播间集合对应有多个第二直播间集合,所述第二直播间集合为与所述第一直播间集合相似的直播间集合;
S102:基于记忆函数,计算获得所述多个第二直播间集合中每个第二直播间集合的再看概率,所述记忆函数用于表示观众在当前时刻对所述第一直播间集合的观看记忆,所述再看概率为在所述第一预设时间范围内观看过所述第一直播间集合又在第二预设时间范围内观看该第二直播间集合的概率;
S103:确定所述多个第二直播间集合中再看概率最大的第二直播间集合;
S104:将所述再看概率最大的第二直播间集合中满足预设条件的直播间推送给所述用户。
具体的,本实施例中的推荐方法可以应用于直播应用程序的服务器端,当服务器确定某一用户对应的客户端开启直播应用程序情况下,服务器中保存了各个用户的历史记录,进而服务器在步骤S101中通过调用与该用户对于的历史记录,确定该用户在第一预设时间范围内观看过的第一直播间集合S。同时,本实施例中的推荐方法可以应用于安装直播应用程序的客户端,该客户端可以是手机、平板电脑、台式电脑等电子设备,也可以是其他电子设备,在此,本申请不做限制。当客户端开启直播应用程序情况下,可向服务器请求该客户端的用户对应的历史记录,进而在步骤S101中确定出该用户在第一预设时间范围内观看过的第一直播间集合s。
在本实施例中,第一预设时间范围可根据实际需要进行设定,比如可以设定为距当前时刻的最接近10天中的前5天,比如:存在距当前时刻最接近的10天,按距当前时刻由远至近包括第1天至第10天,第一预设时间范围为第1天至第5天。统计在第1天至第5天内该用户观看过的第一直播间集合s。第一直播间集合s对应有多个相似的第二直播间集合s*,各个第二直播间集合s*互不相同。
进一步,本实施例中的方法,在步骤S102中,通过采用记忆函数度量用户在观看第一直播间集合后对其相似各个第二直播间集合s*的兴趣变化,即:基于记忆函数度计算得到用户在第一预设时间范围内观看过第一直播间集合又在第二预设时间范围内观看该第二直播间集合s*的再看概率。比如:第一直播间集合s对应有3个相似的第二直播间集合分别为s1*、s2*、s3*。可基于记忆函数获得第二直播间集合s1*的再看概率Ps1*为0.6,第二直播间集合 s2*的再看概率Ps2*为0.8,第二直播间集合s3*的再看概率Ps3*为0.2。其中,第二时间范围在第一时间范围之后,比如:存在距当前时刻最接近的10天,按距当前时刻由远至近包括第1天至第10天,第一预设时间范围为第1天至第5天,第二时间范围为第6天至第10天。在具体实施过程中,第一时间范围与第二时间范围可根据实际需要进行设定,在此,本申请不做限制。
进而,在步骤S103中,确定出再看概率最大的第二直播间集合,比如上述示例中第二直播间集合s1*、第二直播间集合s2*、第二直播间集合s3*中再看概率最高的为第二直播间集合s2*。在挑选出再看概率最高的第二直播间集合后,表明用户再看该第二直播间集合中的直播间的意愿较高,可通过步骤S104从再看概率最高的第二直播间集合中挑选出符合预设条件的直播间推送给该用户,由于考虑了用户兴趣的变化,可以使得为用户推荐的直播间更有效。
进一步,在上述步骤S102中,在计算每个第二直播间集合的再看概率时,需要利用记忆函数,记忆函数对应的公式为m(u,t,s)=md(s)+mu(u,t,s),表示用户u在当前时刻t对第一直播间集合s的观看记忆函数,md(s)=N(s*,T)/N(s),md(s)采取全网用户的历史观看记录进行度量,是只与第一直播间集合s有关的默认记忆函数,反应的是大众普遍对第一直播间集合s的潜在观看可能性。其中,N(s)为在第一预设时间范围内观看第一直播间集合s的总人数,N(s *,T)为在所述第一预设时间范围内观看了所述第一直播集合s的所有用户中在观看后的第二预设时间范围T内观看该第二直播间集合s*的人数,T为记忆时间长度,可根据实际需要进行设定,在此,本申请不做限制。
其中,
Figure PCTCN2018082166-appb-000005
mu(u,t,s)是用户u在当前时刻t对第一直播间集合s的观看记忆,反应的是个人对第一直播间集合s的潜在观看可能性。该公式中,λ为衰减系数,是一个大于0的常数。T1为在所述第二时间范围T内用户u观看所述第一直播间集合s的时间点集合,由于用户对直播间的记忆残留会随着时间衰减,因此观看记忆使用的是一个指数分布来进行估计。
通过上述公式m(u,t,s)=md(s)+mu(u,t,s)可以求得单个用户截止当前时刻对第一直播间集合s的记忆值,进而,可由此求得在当前时刻全网用户对于第一直播间集合s的记忆值为
Figure PCTCN2018082166-appb-000006
即统计在第一时间范围内的所有用户对于第一直播间集合s的记忆值总和。通过这样的方式可获得每个第二直播间集合对应的当前时刻全网用户对于第一直播间集合s的记忆值。
进一步,在计算每个第二直播间集合的再看概率时,需要统计在当前时刻观看该第二直播间集合的人数
Figure PCTCN2018082166-appb-000007
以及在第二时间范围内观看过该第二直播间集合的人数
Figure PCTCN2018082166-appb-000008
与该第二直播间集合对应的再看概率为当前时刻观看该第二直播间集合的人数除以当前时刻全网用户对所述第一直播间集 合的记忆值与在所述第二时间范围内观看过该第二直播间集合的人数的差值。该第二直播间集合的再看概率的公式可以表示为:
Figure PCTCN2018082166-appb-000009
下面以一个完整的示例来对第二直播间集合的再看概率的计算进行详细解释。存在距当前时刻最接近的10天,按距当前时刻由远至近包括第1天至第10天,第一预设时间范围为第1天至第5天,第二时间范围为第6天至第10天。第1天至第5天用户观看了第一直播间集合s,针对第一直播间集合s相似的第二直播间集合s4*,通过第6天至第10天的观看记录,可求得第二直播间集合s4*的再看概率。
首先,统计第1天至第5天观看第一直播间集合s的总人数为10000,在第1天至第5天观看了第一直播间集合s后又在第6天至第10天观看了第二直播间集合s4*的人数为2000,则md(s)=2000/10000=0.2。第1天至第5天观看第一直播间集合s的用户A在当前时刻的前2天和前10天看过第二直播间集合s4*,指数分布的衰减系数为0.1,则mu(u,t,s)=e -2*0.1+e -10*0.1=1.186,表明对于用户A而言对第一直播间集合s的记忆值为1.186。假设全网用户在当前时刻对第一直播间集合s的记忆值之和为20000,在当前时刻有500人观看第二直播间集合s4*,在当前时刻之前的第6天至第10天有19000人观看了第二直播间集合s4*,则第二直播间集合s4*的再看概率为500/(20000-19000)=0.5。
进一步,为了确保统计的历史记录的有效性,在统计观看次数时,需要剔除无效的观看次数,比如:在观看时长小于预设时长(如:1分钟、2分钟等)时,确定该次观看为无效,在统计观看次数时会剔除这样的记录。同时,针对第一直播间集合中每个直播间,用户对其观看时长均大于预设时长(如:1分钟、2分钟等)。
具体的,在本实施例中,在步骤S104中,可采用以下两种直播间推送方式:
第一种:在所述用户观看所述第一直播间集合中的第一直播间后,将所述再看概率最大的第二直播间集合中与所述第一直播间相似度大于第一预设阈值的第二直播间推送给所述用户。
具体的,在本实施例中,如果用户观看完了第一直播间集合中的第一直播间A,由于按前述方式确定出了再看概率最大的第二直播间集合,再看概率最大的第二直播间集合可包括多个与第一直播间A相似的的第二直播间B,从这多个第二直播间B中挑选出与第一直播间A相似度大于第一预设阈值的直播间推送给该用户,第一预设阈值可以设置为0.5、0.6等数值,当然,也可以根据需要设定为其它数值,在此本申请不做限制。
第二种:在所述用户观看所述第一直播间集合中的任意一个直播间后,将所述再看概率最大的第二直播间集合中与所述第一直播间集合中直播间的相似度大于第二预设阈值的第二直播间推送给所述用户。
具体的,在本实施例中,如果用户观看完了第一直播间集合中的任意一个直播间,由于按前述方式确定出了再看概率最大的第二直播间集合,如果第一直播间集合包括直播间A1、直播间A2、直播间A3。再看概率最大的第二直播间集合包括直播间B1、直播间B2、直播间B3、直播间B4、直播间B5,其中,直播间B1与直播间B2为与直播间A1相似的直播间,直播间B1与直播间A1的相似度为0.4,直播间B2与直播间A1的相似度为0.7。直播间B3与直播间B4为与直播间A2相似的直播间,直播间B3与直播间A2的相似度为0.65,直播间B4与直播间A2的相似度为0.5。直播间B5为与直播间A3相似的直播间,直播间B5与直播间A3的相似度为0.75。需要从再看概率最大的第二直播间集合中挑选出与第一直播间集合中直播间的相似度大于第二预设阈值的直播间推送给用户如果第二预设阈值设定为0.6,则需要将直播间B2、直播间B3、直播间B5推荐给用户。
当然,推荐方式不限以上两种,还可以采用其他方式,比如:将再看概率最大的第二直播间集合中所有直播间均推送给用户。在此本申请不做限制。
请参见图2,本发明的第三实施例提供了一种电子设备,所述电子设备包括:
第一确定单元201,用于在直播应用程序开启情况下,确定在第一预设时间范围内用户观看过的第一直播间集合,所述第一直播间集合对应有多个第二直播间集合,所述第二直播间集合为与所述第一直播间集合相似的直播间集合;
计算单元202,用于基于记忆函数,计算获得所述多个第二直播间集合中每个第二直播间集合的再看概率,所述记忆函数用于表示观众在当前时刻对所述第一直播间集合的观看记忆,所述再看概率为在所述第一预设时间范围内观看过所述第一直播间集合又在第二预设时间范围内观看该第二直播间集合的概率;
第二确定单元203,用于确定所述多个第二直播间集合中再看概率最大的第二直播间集合;
推送单元204,用于将所述再看概率最大的第二直播间集合中满足预设条件的直播间推送给所述用户。
具体的,本实施例中电子设备可以是直播应用程序的服务器,当服务器确定某一用户对应的客户端开启直播应用程序情况下,服务器中保存了各个用户 的历史记录,进而服务器通过第一确定单元201调用与该用户对于的历史记录,确定该用户在第一预设时间范围内观看过的第一直播间集合S。同时,本实施例中的电子设备可以是安装直播应用程序的客户端,该客户端可以是手机、平板电脑、台式电脑等电子设备,也可以是其他电子设备,在此,本申请不做限制。当客户端开启直播应用程序情况下,可向服务器请求该客户端的用户对应的历史记录,进而通过第一确定单元201确定出该用户在第一预设时间范围内观看过的第一直播间集合s。
在本实施例中,第一预设时间范围可根据实际需要进行设定,比如可以设定为距当前时刻的最接近10天中的前5天,比如:存在距当前时刻最接近的10天,按距当前时刻由远至近包括第1天至第10天,第一预设时间范围为第1天至第5天。统计在第1天至第5天内该用户观看过的第一直播间集合s。第一直播间集合s对应有多个相似的第二直播间集合s*,各个第二直播间集合s*互不相同。
进一步,本实施例中的电子设备,通过计算单元202,采用记忆函数度量用户在观看第一直播间集合后对其相似各个第二直播间集合s*的兴趣变化,即:基于记忆函数度计算得到用户在第一预设时间范围内观看过第一直播间集合又在第二预设时间范围内观看该第二直播间集合s*的再看概率。比如:第一直播间集合s对应有3个相似的第二直播间集合分别为s1*、s2*、s3*。可基于记忆函数获得第二直播间集合s1*的再看概率Ps1*为0.6,第二直播间集合s2*的再看概率Ps2*为0.8,第二直播间集合s3*的再看概率Ps3*为0.2。其中,第二时间范围在第一时间范围之后,比如:存在距当前时刻最接近的10天,按距当前时刻由远至近包括第1天至第10天,第一预设时间范围为第1天至第5天,第二时间范围为第6天至第10天。在具体实施过程中,第一时间范围与第二时间范围可根据实际需要进行设定,在此,本申请不做限制。
进而,通过第二确定单元203确定出再看概率最大的第二直播间集合,比如上述示例中第二直播间集合s1*、第二直播间集合s2*、第二直播间集合s3*中再看概率最高的为第二直播间集合s2*。在挑选出再看概率最高的第二直播间集合后,表明用户再看该第二直播间集合中的直播间的意愿较高,可通过推送单元204从再看概率最高的第二直播间集合中挑选出符合预设条件的直播间推送给该用户,由于考虑了用户兴趣的变化,可以使得为用户推荐的直播间更有效。
进一步,计算单元202在计算每个第二直播间集合的再看概率时,需要利用记忆函数,记忆函数对应的公式为m(u,t,s)=md(s)+mu(u,t,s),表示用户u在当前时刻t对第一直播间集合s的观看记忆函数,md(s)=N(s*,T)/N(s),md(s)采 取全网用户的历史观看记录进行度量,是只与第一直播间集合s有关的默认记忆函数,反应的是大众普遍对第一直播间集合s的潜在观看可能性。其中,N(s)为在第一预设时间范围内观看第一直播间集合s的总人数,N(s *,T)为在所述第一预设时间范围内观看了所述第一直播集合s的所有用户中在观看后的第二预设时间范围T内观看该第二直播间集合s*的人数,T为记忆时间长度,可根据实际需要进行设定,在此,本申请不做限制。
其中,
Figure PCTCN2018082166-appb-000010
mu(u,t,s)是用户u在当前时刻t对第一直播间集合s的观看记忆,反应的是个人对第一直播间集合s的潜在观看可能性。该公式中,λ为衰减系数,是一个大于0的常数。T1为在所述第二时间范围T内用户u观看所述第一直播间集合s的时间点集合,由于用户对直播间的记忆残留会随着时间衰减,因此观看记忆使用的是一个指数分布来进行估计。
通过上述公式m(u,t,s)=md(s)+mu(u,t,s)可以求得单个用户截止当前时刻对第一直播间集合s的记忆值,进而,可由此求得在当前时刻全网用户对于第一直播间集合s的记忆值为
Figure PCTCN2018082166-appb-000011
即统计在第一时间范围内的所有用户对于第一直播间集合s的记忆值总和。通过这样的方式可获得每个第二直播间集合对应的当前时刻全网用户对于第一直播间集合s的记忆值。
进一步,在计算每个第二直播间集合的再看概率时,还需要统计在当前时刻观看该第二直播间集合的人数
Figure PCTCN2018082166-appb-000012
以及在第二时间范围内观看过该第二直播间集合的人数
Figure PCTCN2018082166-appb-000013
与该第二直播间集合对应的再看概率为当前时刻观看该第二直播间集合的人数除以当前时刻全网用户对所述第一直播间集合的记忆值与在所述第二时间范围内观看过该第二直播间集合的人数的差值。第二直播间集合的再看概率的公式可以表示为:
Figure PCTCN2018082166-appb-000014
下面以一个完整的示例来对第二直播间集合的再看概率的计算进行详细解释。存在距当前时刻最接近的10天,按距当前时刻由远至近包括第1天至第10天,第一预设时间范围为第1天至第5天,第二时间范围为第6天至第10天。第1天至第5天用户观看了第一直播间集合s,针对第一直播间集合s相似的第二直播间集合s4*,通过第6天至第10天的观看记录,可求得第二直播间集合s4*的再看概率。
首先,统计第1天至第5天观看第一直播间集合s的总人数为10000,在第1天至第5天观看了第一直播间集合s后又在第6天至第10天观看了第二直播间集合s4*的人数为2000,则md(s)=2000/10000=0.2。第1天至第5天观看第一直播间集合s的用户A在当前时刻的前2天和前10天看过第二直播间集合s4*,指数分布的衰减系数为0.1,则mu(u,t,s)=e -2*0.1+e -10*0.1=1.186,表明对 于用户A而言对第一直播间集合s的记忆值为1.186。假设全网用户在当前时刻对第一直播间集合s的记忆值之和为20000,在当前时刻有500人观看第二直播间集合s4*,在当前时刻之前的第6天至第10天有19000人观看了第二直播间集合s4*,则第二直播间集合s4*的再看概率为500/(20000-19000)=0.5。
进一步,为了确保统计的历史记录的有效性,在统计观看次数时,需要剔除无效的观看次数,比如:在观看时长小于预设时长(如:1分钟、2分钟等)时,确定该次观看为无效,在统计观看次数时会剔除这样的记录。同时,针对第一直播间集合中每个直播间,用户对其观看时长均大于预设时长(如:1分钟、2分钟等)。
进一步,推送单元204可采用以下两种方式进行直播间推送:
第一种:在所述用户观看所述第一直播间集合中的第一直播间后,将所述再看概率最大的第二直播间集合中与所述第一直播间相似度大于第一预设阈值的第二直播间推送给所述用户。
具体的,在本实施例中,如果用户观看完了第一直播间集合中的第一直播间A,由于按前述方式确定出了再看概率最大的第二直播间集合,再看概率最大的第二直播间集合可包括多个与第一直播间A相似的的第二直播间B,从这多个第二直播间B中挑选出与第一直播间A相似度大于第一预设阈值的直播间推送给该用户,第一预设阈值可以设置为0.5、0.6等数值,当然,也可以根据需要设定为其它数值,在此本申请不做限制。
第二种:在所述用户观看所述第一直播间集合中的任意一个直播间后,将所述再看概率最大的第二直播间集合中与所述第一直播间集合中直播间的相似度大于第二预设阈值的第二直播间推送给所述用户。
具体的,在本实施例中,如果用户观看完了第一直播间集合中的任意一个直播间,由于按前述方式确定出了再看概率最大的第二直播间集合,如果第一直播间集合包括直播间A1、直播间A2、直播间A3。再看概率最大的第二直播间集合包括直播间B1、直播间B2、直播间B3、直播间B4、直播间B5,其中,直播间B1与直播间B2为与直播间A1相似的直播间,直播间B1与直播间A1的相似度为0.4,直播间B2与直播间A1的相似度为0.7。直播间B3与直播间B4为与直播间A2相似的直播间,直播间B3与直播间A2的相似度为0.65,直播间B4与直播间A2的相似度为0.5。直播间B5为与直播间A3相似的直播间,直播间B5与直播间A3的相似度为0.75。需要从再看概率最大的第二直播间集合中挑选出与第一直播间集合中直播间的相似度大于第二预设阈值的直播间推送给用户如果第二预设阈值设定为0.6,则需要将直播间B2、直播间B3、直播间B5推荐给用户。
当然,推荐方式不限以上两种,还可以采用其他方式,比如:将再看概率最大的第二直播间集合中所有直播间均推送给用户。在此本申请不做限制。
请参见图3,本发明的第三实施例提供了一种电子设备,该实施例的电子设备包括:处理器301、存储器302以及存储在所述存储器中并可在所述处理器上运行的计算机程序,例如第一实施例中直播间推荐方法对应的程序。所述处理器执行所述计算机程序时实现上述第一实施例中各路径检测中的步骤。或者,所述处理器执行所述计算机程序时实现上述第二实施例的电子设备中各模块/单元的功能。
示例性的,所述计算机程序可以被分割成一个或多个模块/单元,所述一个或者多个模块/单元被存储在所述存储器中,并由所述处理器执行,以完成本发明。所述一个或多个模块/单元可以是能够完成特定功能的一系列计算机程序指令段,该指令段用于描述所述计算机程序在所述计算机装置中的执行过程。例如,所述计算机程序可以被分割成第一确定单元、计算单元、第二确定单元和推送单元,各单元具体功能如下:
第一确定单元,用于在直播应用程序开启情况下,确定在第一预设时间范围内用户观看过的第一直播间集合,所述第一直播间集合对应有多个第二直播间集合,所述第二直播间集合为与所述第一直播间集合相似的直播间集合;
计算单元,用于基于记忆函数,计算获得所述多个第二直播间集合中每个第二直播间集合的再看概率,所述记忆函数用于表示观众在当前时刻对所述第一直播间集合的观看记忆,所述再看概率为在所述第一预设时间范围内观看过所述第一直播间集合又在第二预设时间范围内观看该第二直播间集合的概率;
第二确定单元,用于确定所述多个第二直播间集合中再看概率最大的第二直播间集合;
推送单元,用于将所述再看概率最大的第二直播间集合中满足预设条件的直播间推送给所述用户。
所述电子设备可包括,但不仅限于,处理器、存储器。本领域技术人员可以理解,所述示意图3仅仅是计算机装置的示例,并不构成对电子设备的限定,可以包括比图示更多或更少的部件,或者组合某些部件,或者不同的部件,例如所述电子设备还可以包括输入输出设备、网络接入设备、总线等。
所称处理器301可以是中央处理单元(Central Processing Unit,CPU),还可以是其他通用处理器、数字信号处理器(Digital Signal Processor,DSP)、专用集成电路(Application Specific Integrated Circuit,ASIC)、现成可编程门阵列(Field-Programmable Gate Array,FPGA)或者其他可编程逻辑器件、分立门或者晶体管逻辑器件、分立硬件组件等。通用处理器可以是微处理器或者该处理 器也可以是任何常规的处理器等,所述处理器是所述计算机装置的控制中心,利用各种接口和线路连接整个计算机装置的各个部分。
所述存储器302可用于存储所述计算机程序和/或模块,所述处理器通过运行或执行存储在所述存储器内的计算机程序和/或模块,以及调用存储在存储器内的数据,实现所述计算机装置的各种功能。所述存储器可主要包括存储程序区和存储数据区,其中,存储程序区可存储操作系统、至少一个功能所需的应用程序(比如声音播放功能、图像播放功能等)等;存储数据区可存储根据手机的使用所创建的数据(比如音频数据、视频数据等)等。此外,存储器可以包括高速随机存取存储器,还可以包括非易失性存储器,例如硬盘、内存、插接式硬盘,智能存储卡(Smart Media Card,SMC),安全数字(Secure Digital,SD)卡,闪存卡(Flash Card)、至少一个磁盘存储器件、闪存器件、或其他易失性固态存储器件。
进一步,所述历史时间范围包括距当前时刻最近的N天,N为大于0的整数该电子设备所包括的处理器301还具有以下功能:
在计算所述多个第二直播间集合中每个第二直播间集合的再看概率时,基于记忆函数公式,确定在当前时刻全网用户对所述第一直播间集合的记忆值,以及确定在当前时刻观看该第二直播间集合的人数,以及确定在所述第二时间范围内观看过该第二直播间集合的人数,该第二直播间集合的再看概率为所述当前时刻观看该第二直播间集合的人数除以所述当前时刻全网用户对所述第一直播间集合的记忆值与所述在所述第二时间范围内观看过该第二直播间集合的人数的差值;
其中,所述记忆函数公式为m(u,t,s)=md(s)+mu(u,t,s),表示用户u在当前时刻t对第一直播间集合s的观看记忆函数,md(s)=N(s*,T)/N(s),N(s)为在第一预设时间范围内观看第一直播间集合s的总人数,N(s *,T)为在所述第一预设时间范围内观看了所述第一直播集合s的所有用户中在观看后的第二预设时间范围T内观看该第二直播间集合s*的人数,
Figure PCTCN2018082166-appb-000015
λ为衰减系数,T1为在所述第二时间范围内用户u观看所述第一直播间集合s的时间点集合,所述当前时刻全网用户对所述第一直播间集合的记忆值为
Figure PCTCN2018082166-appb-000016
进一步,该电子设备所包括的处理器301还具有以下功能:
在所述用户观看所述第一直播间集合中的第一直播间后,将所述再看概率最大的第二直播间集合中与所述第一直播间相似度大于第一预设阈值的第二直播间推送给所述用户。
进一步,该电子设备所包括的处理器301还具有以下功能:
在所述用户观看所述第一直播间集合中的任意一个直播间后,将所述再看 概率最大的第二直播间集合中与所述第一直播间集合中直播间的相似度大于第二预设阈值的第二直播间推送给所述用户。
进一步,所述第一直播间集合中的每个直播间的观看时长均大于预设时长。
本发明第四实施例提供了一种计算机可读存储介质,其上存储有计算机程序,本发明第二实施例中的所述电子设备集成的功能单元如果以软件功能单元的形式实现并作为独立的产品销售或使用时,可以存储在一个计算机可读取存储介质中。基于这样的理解,本发明实现上述第一实施例的直播间推荐方法中的全部或部分流程,也可以通过计算机程序来指令相关的硬件来完成,所述的计算机程序可存储于一计算机可读存储介质中,该计算机程序在被处理器执行时,可实现上述各个方法实施例的步骤。其中,所述计算机程序包括计算机程序代码,所述计算机程序代码可以为源代码形式、对象代码形式、可执行文件或某些中间形式等。所述计算机可读介质可以包括:能够携带所述计算机程序代码的任何实体或装置、记录介质、U盘、移动硬盘、磁碟、光盘、计算机存储器、只读存储器(ROM,Read-Only Memory)、随机存取存储器(RAM,Random Access Memory)、电载波信号、电信信号以及软件分发介质等。需要说明的是,所述计算机可读介质包含的内容可以根据司法管辖区内立法和专利实践的要求进行适当的增减,例如在某些司法管辖区,根据立法和专利实践,计算机可读介质不包括电载波信号和电信信号。
在本发明实施例的技术方案中,在直播应用程序开启情况下,确定在第一预设时间范围内用户观看过的第一直播间集合,该第一直播间集合包括多个相似的第二直播间集合。进而,通过记忆函数来准确度量用户在观看第一直播间集合后再观看相似的第二直播间集合的再看概率,当计算好每个相似的第二直播间集合的再看概率后,挑选出再看概率最大的第二直播间集合,将再看概率最大的第二直播间集合中满足预设条件的直播间推送给用户。这样,通过记忆函数来准确度量用户在观看直播间后对其相似直播间的兴趣变化,充分考虑了用户的兴趣变化,为用户提供有效且可靠的推荐直播间。
尽管已描述了本发明的优选实施例,但本领域内的技术人员一旦得知了基本创造性概念,则可对这些实施例作出另外的变更和修改。所以,所附权利要求意欲解释为包括优选实施例以及落入本发明范围的所有变更和修改。
显然,本领域的技术人员可以对本发明进行各种改动和变型而不脱离本发明的精神和范围。这样,倘若本发明的这些修改和变型属于本发明权利要求及其等同技术的范围之内,则本发明也意图包含这些改动和变型在内。

Claims (10)

  1. 一种直播间推送方法,其特征在于,包括:
    在直播应用程序开启情况下,确定在第一预设时间范围内用户观看过的第一直播间集合,所述第一直播间集合对应有多个第二直播间集合,所述第二直播间集合为与所述第一直播间集合相似的直播间集合;
    基于记忆函数,计算获得所述多个第二直播间集合中每个第二直播间集合的再看概率,所述记忆函数用于表示观众在当前时刻对所述第一直播间集合的观看记忆,所述再看概率为在所述第一预设时间范围内观看过所述第一直播间集合又在第二预设时间范围内观看该第二直播间集合的概率;
    确定所述多个第二直播间集合中再看概率最大的第二直播间集合;
    将所述再看概率最大的第二直播间集合中满足预设条件的直播间推送给所述用户。
  2. 如权利要求1所述的方法,其特征在于,所述基于记忆函数,计算获得所述多个第二直播间集合中每个第二直播间集合的再看概率,包括:
    在计算所述多个第二直播间集合中每个第二直播间集合的再看概率时,基于记忆函数公式,确定在当前时刻全网用户对所述第一直播间集合的记忆值,以及确定在当前时刻观看该第二直播间集合的人数,以及确定在所述第二时间范围内观看过该第二直播间集合的人数,该第二直播间集合的再看概率为所述当前时刻观看该第二直播间集合的人数除以所述当前时刻全网用户对所述第一直播间集合的记忆值与所述在所述第二时间范围内观看过该第二直播间集合的人数的差值;
    其中,所述记忆函数公式为m(u,t,s)=md(s)+mu(u,t,s),表示用户u在当前时刻t对第一直播间集合s的观看记忆函数,md(s)=N(s*,T)/N(s),N(s)为在第一预设时间范围内观看第一直播间集合s的总人数,N(s *,T)为在所述第一预设时间范围内观看了所述第一直播集合s的所有用户中在观看后的第二预设时间范围T内观看该第二直播间集合s*的人数,
    Figure PCTCN2018082166-appb-100001
    λ为衰减系数,T1为在所述第二时间范围内用户u观看所述第一直播间集合s的时间点集合,所述当前时刻全网用户对所述第一直播间集合的记忆值为
    Figure PCTCN2018082166-appb-100002
  3. 如权利要求1所述的方法,其特征在于,所述将所述再看概率最大的第二直播间集合中满足预设条件的直播间推送给所述用户,包括:
    在所述用户观看所述第一直播间集合中的第一直播间后,将所述再看概率最大的第二直播间集合中与所述第一直播间相似度大于第一预设阈值的第二直播间推送给所述用户。
  4. 如权利要求1所述的方法,其特征在于,所述将所述再看概率最大的第二直播间集合中满足预设条件的直播间推送给所述用户,包括:
    在所述用户观看所述第一直播间集合中的任意一个直播间后,将所述再看概率最大的第二直播间集合中与所述第一直播间集合中直播间的相似度大于第二预设阈值的第二直播间推送给所述用户。
  5. 如权利要求1所述的方法,其特征在于,所述第一直播间集合中的每个直播间的观看时长均大于预设时长。
  6. 一种电子设备,其特征在于,包括:
    第一确定单元,用于在直播应用程序开启情况下,确定在第一预设时间范围内用户观看过的第一直播间集合,所述第一直播间集合对应有多个第二直播间集合,所述第二直播间集合为与所述第一直播间集合相似的直播间集合;
    计算单元,用于基于记忆函数,计算获得所述多个第二直播间集合中每个第二直播间集合的再看概率,所述记忆函数用于表示观众在当前时刻对所述第一直播间集合的观看记忆,所述再看概率为在所述第一预设时间范围内观看过所述第一直播间集合又在第二预设时间范围内观看该第二直播间集合的概率;
    第二确定单元,用于确定所述多个第二直播间集合中再看概率最大的第二直播间集合;
    推送单元,用于将所述再看概率最大的第二直播间集合中满足预设条件的直播间推送给所述用户。
  7. 如权利要求6所述的电子设备,其特征在于,所述计算单元用于:
    在计算所述多个第二直播间集合中每个第二直播间集合的再看概率时,基于记忆函数公式,确定在当前时刻全网用户对所述第一直播间集合的记忆值,以及确定在当前时刻观看该第二直播间集合的人数,以及确定在所述第二时间范围内观看过该第二直播间集合的人数,该第二直播间集合的再看概率为所述当前时刻观看该第二直播间集合的人数除以所述当前时刻全网用户对所述第一直播间集合的记忆值与所述在所述第二时间范围内观看过该第二直播间集合的人数的差值;
    其中,所述记忆函数公式为m(u,t,s)=md(s)+mu(u,t,s),表示用户u在当前时刻t对第一直播间集合s的观看记忆函数,md(s)=N(s*,T)/N(s),N(s)为在第一预设时间范围内观看第一直播间集合s的总人数,N(s *,T)为在所述第一预设时间范围内观看了所述第一直播集合s的所有用户中在观看后的第二预设时间范围T内观看该第二直播间集合s*的人数,
    Figure PCTCN2018082166-appb-100003
    λ为衰减系数,T1为在所述第二时间范围内用户u观看所述第一直播间集合s的时间点集合,所述当前时刻全网用户对所述第一直播间集合的记忆值为
    Figure PCTCN2018082166-appb-100004
  8. 如权利要求7所述的电子设备,其特征在于,所述推送单元用于:
    在所述用户观看所述第一直播间集合中的任意一个直播间后,将所述再看概率最大的第二直播间集合中与所述第一直播间集合中直播间的相似度大于第二预设阈值的第二直播间推送给所述用户。
  9. 一种电子设备,其特征在于,所述电子设备包括处理器,所述处理器用于在执行存储器中存储的计算机程序时实现如权利要求1-5中任意项所述的直播间推荐方法中的步骤。
  10. 一种可读存储介质,其上存储有计算机程序,其特征在于,在所述计算机程序被处理器执行时实现如权利要求1-5中任意项所述的直播间推荐方法中的步骤。
PCT/CN2018/082166 2018-01-04 2018-04-08 一种直播间推荐方法、电子设备及可读存储介质 WO2019134285A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201810008939.0A CN108235045B (zh) 2018-01-04 2018-01-04 一种直播间推荐方法、电子设备及可读存储介质
CN201810008939.0 2018-01-04

Publications (1)

Publication Number Publication Date
WO2019134285A1 true WO2019134285A1 (zh) 2019-07-11

Family

ID=62645337

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2018/082166 WO2019134285A1 (zh) 2018-01-04 2018-04-08 一种直播间推荐方法、电子设备及可读存储介质

Country Status (2)

Country Link
CN (1) CN108235045B (zh)
WO (1) WO2019134285A1 (zh)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111294607A (zh) * 2020-01-22 2020-06-16 北京达佳互联信息技术有限公司 直播互动方法、装置、服务器及终端
CN112261423A (zh) * 2020-10-16 2021-01-22 北京百度网讯科技有限公司 用于推送信息的方法、装置、设备以及存储介质

Families Citing this family (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109151488B (zh) * 2018-07-06 2021-07-23 武汉斗鱼网络科技有限公司 根据用户行为实时推荐直播间的方法及系统
CN108989893B (zh) * 2018-08-21 2020-12-15 武汉斗鱼网络科技有限公司 一种直播间推荐方法、装置、终端和存储介质
CN110895594A (zh) * 2018-08-23 2020-03-20 武汉斗鱼网络科技有限公司 一种页面展示的方法以及相关设备
CN109379608B (zh) * 2018-09-13 2021-07-23 武汉斗鱼网络科技有限公司 一种直播间的推荐方法以及相关设备
CN109146581A (zh) * 2018-09-30 2019-01-04 武汉斗鱼网络科技有限公司 一种资源分配方法、装置及可读存储介质
CN110087119B (zh) * 2019-04-26 2022-02-22 广州酷狗计算机科技有限公司 直播首页显示方法、装置及计算机可读存储介质
CN113473160B (zh) * 2020-03-31 2023-05-02 北京达佳互联信息技术有限公司 一种直播间的匹配方法、装置、设备及存储介质

Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1624684A (zh) * 2003-12-02 2005-06-08 索尼株式会社 信息处理器、信息处理方法和计算机程序
CN102957949A (zh) * 2012-05-18 2013-03-06 华东师范大学 为用户推荐视频的装置及方法
US9055343B1 (en) * 2013-06-07 2015-06-09 Google Inc. Recommending content based on probability that a user has interest in viewing the content again
CN105847984A (zh) * 2016-03-25 2016-08-10 乐视控股(北京)有限公司 一种视频推荐方法及装置
CN106570090A (zh) * 2016-10-20 2017-04-19 杭州电子科技大学 基于兴趣变化和信任关系的协同过滤推荐方法
CN106791966A (zh) * 2016-12-28 2017-05-31 武汉斗鱼网络科技有限公司 一种基于改进型关联规则的直播间推荐方法及系统
CN107205178A (zh) * 2017-04-25 2017-09-26 北京潘达互娱科技有限公司 直播间推荐方法及装置

Patent Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1624684A (zh) * 2003-12-02 2005-06-08 索尼株式会社 信息处理器、信息处理方法和计算机程序
CN102957949A (zh) * 2012-05-18 2013-03-06 华东师范大学 为用户推荐视频的装置及方法
US9055343B1 (en) * 2013-06-07 2015-06-09 Google Inc. Recommending content based on probability that a user has interest in viewing the content again
CN105847984A (zh) * 2016-03-25 2016-08-10 乐视控股(北京)有限公司 一种视频推荐方法及装置
CN106570090A (zh) * 2016-10-20 2017-04-19 杭州电子科技大学 基于兴趣变化和信任关系的协同过滤推荐方法
CN106791966A (zh) * 2016-12-28 2017-05-31 武汉斗鱼网络科技有限公司 一种基于改进型关联规则的直播间推荐方法及系统
CN107205178A (zh) * 2017-04-25 2017-09-26 北京潘达互娱科技有限公司 直播间推荐方法及装置

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111294607A (zh) * 2020-01-22 2020-06-16 北京达佳互联信息技术有限公司 直播互动方法、装置、服务器及终端
CN111294607B (zh) * 2020-01-22 2021-10-08 北京达佳互联信息技术有限公司 直播互动方法、装置、服务器及终端
US11825134B2 (en) 2020-01-22 2023-11-21 Beijing Dajia Internet Information Technology Co., Ltd. Method for interacting in live-streaming and server
CN112261423A (zh) * 2020-10-16 2021-01-22 北京百度网讯科技有限公司 用于推送信息的方法、装置、设备以及存储介质

Also Published As

Publication number Publication date
CN108235045B (zh) 2019-09-10
CN108235045A (zh) 2018-06-29

Similar Documents

Publication Publication Date Title
WO2019134285A1 (zh) 一种直播间推荐方法、电子设备及可读存储介质
WO2019134290A1 (zh) 一种直播间排序方法、电子设备及可读存储介质
US10945043B2 (en) Methods and apparatus to create a panel of media device users
US9813309B2 (en) Method, apparatus, and system for determining a location corresponding to an IP address
US11030190B2 (en) Experience analytic objects, systems and methods
CN108322829A (zh) 个性化主播推荐方法、装置及电子设备
US20150289021A1 (en) System and method for collecting viewer opinion information
US10140364B1 (en) Dynamically altering shared content
WO2018045893A1 (zh) 媒体文件推送方法、媒体文件服务器及媒体文件推送系统
US9367861B2 (en) Timer-based ad placement in content retrieval applications
CN107257986B (zh) 动态内容显示时间调整
WO2018133566A1 (zh) 推广信息新鲜度确定方法及装置、电子设备及存储介质
WO2017156984A1 (zh) 一种移动终端的视频更新推送方法以及客户端
CN109428910B (zh) 一种数据处理方法、装置及系统
WO2019165697A1 (zh) 刷人气用户的识别方法、装置、终端设备及储存介质
CN110362458B (zh) 应用程序评价提示方法、装置、电子设备及可读存储介质
CN114747227A (zh) 跨人口统计群体估计普查级受众规模和总印象持续时间的方法、系统和装置
CN111523920B (zh) 一种信息推送方法、装置及终端设备
CN109688217B (zh) 一种消息推送方法、装置及电子设备
US8990358B2 (en) Systems and methods for expedited delivery of media content
CN105956061B (zh) 一种用户间相似度确定方法及装置
CN110324713A (zh) 虚拟资源的管理方法及装置、直播系统
CN111083534B (zh) 一种用于提供推荐视频列表的方法与设备
CN112785333A (zh) 广告投放方法、电子设备、存储介质及装置
WO2018205642A1 (zh) 视频收益计算建模装置、方法和视频推荐装置、方法和服务器及存储介质

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 18898511

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 18898511

Country of ref document: EP

Kind code of ref document: A1