CN105163142A - User preference determination method, video recommendation method, user preference determination system and video recommendation system - Google Patents
User preference determination method, video recommendation method, user preference determination system and video recommendation system Download PDFInfo
- Publication number
- CN105163142A CN105163142A CN201510570624.1A CN201510570624A CN105163142A CN 105163142 A CN105163142 A CN 105163142A CN 201510570624 A CN201510570624 A CN 201510570624A CN 105163142 A CN105163142 A CN 105163142A
- Authority
- CN
- China
- Prior art keywords
- video
- user
- mrow
- time length
- msub
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 61
- 238000004364 calculation method Methods 0.000 claims description 19
- 238000012545 processing Methods 0.000 claims description 13
- 238000010586 diagram Methods 0.000 description 16
- 230000006399 behavior Effects 0.000 description 13
- 239000011159 matrix material Substances 0.000 description 10
- 238000006243 chemical reaction Methods 0.000 description 9
- 230000006870 function Effects 0.000 description 7
- 238000004590 computer program Methods 0.000 description 5
- 238000012986 modification Methods 0.000 description 5
- 230000004048 modification Effects 0.000 description 5
- 230000001413 cellular effect Effects 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 3
- 230000008569 process Effects 0.000 description 3
- 230000005540 biological transmission Effects 0.000 description 2
- 238000004891 communication Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 238000000605 extraction Methods 0.000 description 2
- 230000002123 temporal effect Effects 0.000 description 2
- 230000004075 alteration Effects 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 238000013459 approach Methods 0.000 description 1
- 230000003247 decreasing effect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 238000010606 normalization Methods 0.000 description 1
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
- H04N21/20—Servers specifically adapted for the distribution of content, e.g. VOD servers; Operations thereof
- H04N21/25—Management operations performed by the server for facilitating the content distribution or administrating data related to end-users or client devices, e.g. end-user or client device authentication, learning user preferences for recommending movies
- H04N21/258—Client or end-user data management, e.g. managing client capabilities, user preferences or demographics, processing of multiple end-users preferences to derive collaborative data
- H04N21/25866—Management of end-user data
- H04N21/25891—Management of end-user data being end-user preferences
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
- H04N21/40—Client devices specifically adapted for the reception of or interaction with content, e.g. set-top-box [STB]; Operations thereof
- H04N21/45—Management 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
- H04N21/466—Learning process for intelligent management, e.g. learning user preferences for recommending movies
- H04N21/4667—Processing of monitored end-user data, e.g. trend analysis based on the log file of viewer selections
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
- H04N21/40—Client devices specifically adapted for the reception of or interaction with content, e.g. set-top-box [STB]; Operations thereof
- H04N21/45—Management 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
- H04N21/466—Learning process for intelligent management, e.g. learning user preferences for recommending movies
- H04N21/4668—Learning process for intelligent management, e.g. learning user preferences for recommending movies for recommending content, e.g. movies
Landscapes
- Engineering & Computer Science (AREA)
- Databases & Information Systems (AREA)
- Multimedia (AREA)
- Signal Processing (AREA)
- Computer Graphics (AREA)
- Two-Way Televisions, Distribution Of Moving Picture Or The Like (AREA)
Abstract
The invention discloses a user preference determination method, a video recommendation method, a user preference determination system and a video recommendation system. The user preference determination method comprises the steps of: obtaining a video record within a set time length to obtain a video collection and a user collection, wherein videos in the video collection are the videos which are watched within the set time length, and users in the user collection are the users who watched the videos within the set time length; calculating a first average time length value of the videos which are watched by the u-th user within the set time length and calculating a second average time length value of the v-th video which is watched by all the users within the set time length; calculating a preference value of the u-th user for the v-th video based on the time length value of the v-th video watched by the u-th user within the set time length, the first average time length value and the second average time length value to obtain a user preference record, wherein the user preference record contains user identifiers, video identifiers and preference values. The method and system realize user preference determination based on video records of the videos watched by the users.
Description
Technical Field
The invention relates to the technical field of multimedia, in particular to a user preference determination method, a video recommendation method and a video recommendation system.
Background
With the development of video services, as a video platform, it is increasingly important to handle the inter-solving relationship between massive video resources and huge user groups. In order to solve the contradiction between the rapid increase of users and video resources, video recommendation technology is developed. The video recommendation technology is to obtain an interest model of a user by analyzing behaviors of the user on video resources, so as to find video resources that the user may like and recommend the video resources to the user.
In the current video recommendation service of the smart television, video recommendation is mainly based on extraction of user behaviors. The extraction of the user behaviors is to map various behaviors of the user into the preference degree of the user on the video. The behaviors of the user comprise video watching, video scoring, video browsing and other operations, the behaviors substantially reflect the preference of the user on the video, and the user can record the video by using a video platform such as a smart television.
Therefore, how to select and utilize the recorded user behaviors to obtain the video preference of the user so as to further improve the inter-request relationship between the user and the video resource is a direction to be researched in the technical field of video recommendation.
Disclosure of Invention
The embodiment of the invention provides a user preference determining method, a video recommending method and a video recommending system, which are used for determining the preference of a user for a video based on a video record formed by the user watching the video.
The method for determining the user preference based on the video recording implementation provided by one embodiment of the invention comprises the following steps:
acquiring video records within a set time length to obtain a video set and a user set, wherein videos in the video set are videos watched within the set time length, users in the user set are users who have watched within the set time length, the video set comprises M videos, the user set comprises N users, and M, N are integers greater than or equal to 1;
according to the obtained video record, aiming at the u-th user in the user set, calculating a first average time length value of the u-th user watching the video within the set time length, wherein u is more than or equal to 1 and less than or equal to N;
according to the obtained video record, aiming at the v-th video in the video set, calculating a second average time length value of the v-th video watched by all users watching the v-th video within the set time length, wherein v is more than or equal to 1 and less than or equal to M;
and calculating the preference value of the u user to the v video according to the time length value of the v video watched by the u user in the set time length, the first average time length value and the second average time length value to obtain a user preference record, wherein the user preference record comprises a user identifier, a video identifier and a preference value.
One embodiment of the present invention provides a user preference determination system implemented based on video recording, including:
the acquisition module is used for acquiring video records within a set time length to obtain a video set and a user set, wherein videos in the video set are videos watched within the set time length, users in the user set are users who have watched within the set time length, the video set comprises M videos, the user set comprises N users, and M, N are integers greater than or equal to 1;
the first calculation module is used for calculating a first average time length value of the u user watching the video within the set time length according to the obtained video record, wherein u is more than or equal to 1 and less than or equal to N; the video acquisition unit is used for acquiring a video set of a user, and calculating a second average time length value of all users watching the v-th video in the set time length aiming at the v-th video in the video set according to the acquired video record, wherein v is more than or equal to 1 and less than or equal to M;
and the second calculation module is used for calculating the preference value of the u user to the v video according to the time length value of the u user watching the v video in the set time length, the first average time length value and the second average time length value to obtain a user preference record, wherein the user preference record comprises a user identifier, a video identifier and a preference value.
The video recommendation method implemented based on the user preference record obtained by the user preference determination method provided by one embodiment of the invention comprises the following steps:
acquiring a user identifier;
inquiring the user preference record according to the user identification;
and sending video recommendation information to the user equipment according to the query result.
A video recommendation system implemented based on a user preference record obtained by a user preference determination system according to an embodiment of the present invention includes:
the acquisition module is used for acquiring a user identifier;
the query module is used for querying the user preference record according to the user identification;
and the recommendation module is used for sending video recommendation information to the user equipment according to the query result.
In the above embodiment of the present invention, the user preference determination is implemented based on the video record formed by the video watched by the user, that is, first obtaining the video record within the set time length to obtain the video set and the user set, then according to the obtained video record, calculating, for the u-th user in the user set, a first average time length value of the u-th user watching the video within the set time length, calculating, for the v-th video in the video set, a second average time length value of all the users watching the v-th video within the set time length, and finally calculating, according to the time length value of the u-th user watching the v-th video within the set time length, the first average time length value and the second average time length value, the preference value of the u-th user to the v-th video to obtain the user preference record, where the user preference record includes the user identifier, Video identification, preference value. The video recommendation is realized based on the user preference record obtained by the user preference determining method, namely, the user identification is firstly obtained, then the user preference record is inquired according to the user identification, and finally the video recommendation information is sent to the user equipment according to the inquiry result. In the embodiment of the present invention, by obtaining the video record within the set time length, the record includes the user behavior data that does not directly indicate the user preference degree, such as the time length value of the user watching the video, that is, the implicit feedback data of the user, and performing explicit conversion on the implicit feedback data through calculation, the preference value of the user to the video is obtained through calculation, so as to obtain the user preference record, and based on the obtained user preference record, the user preference record can be queried according to the obtained user identifier, so as to perform video recommendation. The embodiment of the invention can be seen in that the user preference is determined based on the video record of the video watched by the user, the video record information reflecting the preference degree of the video watched by the user is utilized, the video record can be directly recorded by the video platform, so that additional network resources are not occupied, the user experience is not influenced, meanwhile, for the video platform, the record belongs to resources easy to obtain, and the record can be stored and accumulated in the video platform, so that a larger amount of watching record of the user can be obtained to determine the user preference record, and the recording is further used in the video recommendation service.
Drawings
In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings needed to be used in the description of the embodiments will be briefly introduced below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and it is obvious for those skilled in the art to obtain other drawings based on these drawings without inventive exercise.
Fig. 1 is a schematic diagram of a network architecture to which the embodiments of the present invention are applicable;
fig. 2 is a schematic flowchart of a user preference determination method implemented based on video recording according to an embodiment of the present invention;
fig. 3 is a schematic flowchart of a video recommendation method implemented based on a user preference record obtained by a user preference determination method according to an embodiment of the present invention;
fig. 4 is a video watching duration value distribution diagram of a user in an actual application scene according to the video recording-based user preference determination method provided in the embodiment of the present invention;
fig. 5 is a first average duration value distribution diagram of the user preference determination method implemented based on video recording in an actual application scenario according to the embodiment of the present invention;
fig. 6 is a second average duration value distribution diagram of the user preference determination method implemented based on video recording in an actual application scenario according to the embodiment of the present invention;
fig. 7 is a histogram of user preference values calculated by the method for determining user preference based on video recording in an actual application scenario according to the embodiment of the present invention;
fig. 8 is a schematic structural diagram of a user preference determining system implemented based on video recording according to an embodiment of the present invention;
fig. 9 is a schematic structural diagram of a video recommendation system implemented based on a user preference record obtained by the user preference determination system according to an embodiment of the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention clearer, the present invention will be described in further detail with reference to the accompanying drawings, and it is apparent that the described embodiments are only a part of the embodiments of the present invention, not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
As the inter-solving relationship between massive video resources and huge user groups becomes more important, a current demand is to obtain an interest model of a user by analyzing the behavior of the user on the video resources, so that a video recommendation service can be provided for the user. The behavior of the user on the video resource may include explicit feedback and implicit feedback, the displayed feedback may be data information such as video scores that directly reflect the video frequency offset of the user, and such feedback often requires direct participation of the user, and therefore additional network resources may be occupied, the experience of the user may be affected, and large-scale data is not easily acquired for video recommendation. The implicit feedback can be data information which does not directly reflect good video frequency offset of the user, such as video watching record of the user, and the like, for the video platform, the implicit feedback belongs to data resources which are easy to obtain in a network, and additional network resources do not need to be occupied. Therefore, how to determine the preference of the user for the video by using the video recording information of the user watching the video, which is easy to obtain, in the video recommendation technology is a technical requirement in the technical field of video recommendation.
In order to solve the above needs, embodiments of the present application provide a method for determining user preference based on video recording, and a method and a system for recommending video based on user preference recording obtained by the method for determining user preference.
In the embodiment of the invention, the user preference record is obtained by obtaining the video record of the user, wherein the record comprises information such as the time length value of the video watched by the user, calculating the first average time length value of all the videos watched by the user in the set time length and calculating the second average time length value of a certain video watched by all the users watching the videos in the set time length, and carrying out explicit conversion on the implicit feedback of the users to obtain the preference value of the user to the video. Based on the user preference record, the video recommendation information can be sent to the user equipment by acquiring the user identifier and carrying out corresponding query according to the user identifier.
Fig. 1 schematically shows a network architecture to which an embodiment of the present invention is applicable.
As shown in fig. 1, the architecture may include: user equipment 101, user preference determination system 102, video recommendation system 103, video recording database 104, user preference recording database 105. The user device 101 and the video recommendation system 103 interact information via the network 106, and the user preference determination system 102 and the video recommendation system 103 can access the video record database 104 and the user preference record database 105.
The network 106 may include an access network, which may include a wireless cellular access network and may also include a wireless local area network, and a convergence/core network, which may also be a wired network, and a backbone transport network (not shown in the figure). The convergence/core network mainly implements transmission convergence, and implements functions such as mobility management in a cellular wireless communication network. The backbone transport network is used for realizing high-speed transmission and exchange of data.
The number of the user equipments 101 may be plural, and the specific number may be determined according to the access capability of the network 106 or the access capability of the application system. The type of user equipment 101 may include a variety of types, as may the manner of access to the network 106. For example, the user equipment 101 may be a device having a video playing function, such as a PC (personal computer), and may be accessed through a wired manner. The user equipment 101 may also include a terminal with a video playing function, which can perform wireless communication, such as a network television platform, a mobile phone platform, and the like, and is accessed through a wireless access manner (such as an access network through a cellular wireless network, or a wireless local area network, and the like).
User preference determination system 102 may obtain information from video recording database 104 and store information into user preference recording database 105; the user equipment 101 can send a video service request carrying a user identifier to the video recommendation system 103 through the network 106; the video recommendation system 103 may obtain information from the user preference record database 105 and provide video recommendation services to the user devices 101 over the network 106.
In practical applications, the user preference determining system 102 and the video recommending system 103 may be independent software systems, or may be functional modules integrated in other application systems, for example, functional modules integrated in a video-on-demand system.
The user preference determining system 102 and the video recommending system 103 are only logically distinguished as different systems, and the hardware implementation of the user preference determining system 102 and the video recommending system 103 is not limited in the embodiment of the present application, for example, the user preference determining system 102 and the video recommending system 103 may be integrated in one device on hardware, or may be implemented by a computer cluster.
It should be noted that, the user identifier used for determining the user in the embodiment of the present application may be a login account of a video platform, such as a login account of a network television; the account information may also be account information registered in an ISP (internet service provider), such as a mailbox account, a microblog account, a mobile phone number, or other user account information, and the user identifier is not limited in this embodiment.
Embodiments of the present invention will be described in detail below with reference to the accompanying drawings.
Fig. 2 is a schematic flowchart of a user preference determination method implemented based on video recording according to an embodiment of the present invention, where the flowchart may be implemented by a user preference determination system, and the flowchart includes the following steps:
step 201: the method comprises the steps of obtaining video records within a set time length to obtain a video set and a user set, wherein videos in the video set are videos watched within the set time length, users in the user set are users who have watched the videos within the set time length, the video set comprises M videos, the user set comprises N users, and M, N are integers greater than or equal to 1.
The video recording refers to a recording formed by a user watching a video, and can record the content of the video watched by the user, such as an identifier, watching time (which may include a watching start time and a watching end time), watching duration and the like. Specifically, the video record may include information such as a user identifier, a video identifier, a timestamp, and a time length value for the user to watch the video. The video recording may be stored in a video recording database.
In particular, the set time length may be preconfigured in the user preference determination system, and embodiments of the present invention also allow setting the time length or modifying the preconfigured time length by manually submitting instructions. The set time period may be a period of one week or one month. When a video viewing record within a set period of time needs to be acquired, the user preference determination system may call the video record within the set period of time from the video record database according to the viewing time in the video record database, including information of the viewing start time and the viewing end time.
In step 201, all video records within a set time period may be acquired, a video record corresponding to the video type within the set time period may be acquired according to the set video type, a video record corresponding to the user type within the set time period may be acquired according to the set user type, or a video record meeting a certain condition within the set time period may be acquired according to a preset policy. For example, in some embodiments, the user group may be divided into a plurality of user groups in the policy, for example, the user groups may be divided into a plurality of groups according to user grades, and when a video watching record of a certain user group needs to be obtained, the video recommendation system may call up, from the video record database, video records of users in the certain user group within a set time length according to the identification information of the user group included in the external instruction. In other embodiments, the video resource may be divided into a plurality of video sets in the policy, for example, into a plurality of sets according to video types, and when a video record of a specific video set needs to be obtained, the video recommendation system may call the video record of the specific video set in the video record database within a set time length according to an identifier (for example, a type identifier) of the specific video set included in the external instruction.
The video record within the set time length is obtained, the obtained video set is a video set observed within the set time length, or a video record corresponding to the video type within the set time length is obtained according to the set video type, so that the obtained video set is obtained. The video set comprises M videos, the element in the video set may be a video identifier, and each video in the video set is respectively assigned with an index number for identifying the position of the video in the video set.
The video records within the set time length are obtained, the obtained user set is the user set which has viewed the video within the set time length, or the video records corresponding to the user type within the set time length are obtained according to the set user type, so that the obtained user set is obtained. The user set includes N users, the elements in the user set may be user identifiers, each user in the user set is respectively assigned with an index number for identifying the position of the user in the user set, and M, N are integers greater than or equal to 1.
For the convenience of subsequent calculation, a video recording matrix can be generated according to the two sets and the time length of the video watched by the user, the matrix can be an M × N matrix, and elements in the matrix are the time length of the corresponding video watched by the corresponding user.
Step 202: and according to the obtained video record, aiming at the u-th user in the user set, calculating a first average time length value of the u-th user watching the video within the set time length, wherein u is more than or equal to 1 and less than or equal to N.
Specifically, a first average duration value of the viewing of the video by the u-th user within the set duration may be calculated according to the following formula:
wherein,represents the first average duration value, | VuL represents the number of videos watched by the u-th user within the set time length, tu,vA value representing a duration of time during which the uth user viewed the vth video within the set period of time,the time length values of all videos watched by the u-th user within the set time length are accumulated.
In specific implementation, according to the mxn video recording matrix generated in step 201, for each column in the matrix, the elements of each row of the matrix corresponding to the column are accumulated to obtain an accumulated sum of each column, where the accumulated sum represents a duration value of all videos watched by the user corresponding to the column within a set duration, that is, a duration value in formula (1)
Step 203: and according to the obtained video record, aiming at the v-th video in the video set, calculating a second average time length value of the v-th video watched by all users watching the v-th video within the set time length, wherein v is more than or equal to 1 and less than or equal to M.
Specifically, the second average duration value of the vth video watched by all users watching the vth video within the set duration may be calculated according to the following formula:
wherein,represents the second average duration value, | UvL represents the number of all users watching the vth video in the set time length, tu,vA value representing a duration of time during which the uth user viewed the vth video within the set period of time,and the time length value of each user watching the v-th video in the set of users watching the v-th video in the set time length is accumulated.
In specific implementation, for each row in the matrix according to the mxn video recording matrix generated in step 201, the elements of each column of the matrix corresponding to the row are accumulated to obtain an accumulated sum of each row, where the accumulated sum represents a total time length value of the video corresponding to the row viewed within a set time length, that is, a total time length value in formula (2)
Step 204: and calculating the preference value of the u user to the v video according to the time length value of the v video watched by the u user in the set time length, the first average time length value and the second average time length value to obtain a user preference record, wherein the user preference record comprises a user identifier, a video identifier and a preference value.
Specifically, logarithmic processing is performed on the duration value of the vth video watched by the u user within the set time length, the first average duration value and the second average duration value respectively; and calculating the preference value of the u user to the v video according to the time length value of the v video watched by the u user in the set time length and the result of logarithmic processing of the first average time length value and the second average time length value.
For example, the distribution of the video watching duration values in the statistical range is taken as an example, the distribution of the video watching duration values presents a long tail distribution characteristic, that is, when the duration values are sorted from large to small, the distribution is a decreasing curve, the duration value is large at the initial part of the curve, then the curve is in a sharp descending trend, and the tail of the curve is not reduced to zero, but is extremely slowly close to the horizontal axis. Therefore, the implicit feedback data of the user represented by the video watching time length value of the user cannot be directly used for calculating the preference value of the user, and a reasonable explicit conversion method is needed to reduce the influence caused by too large difference of the watching time length magnitude.
According to the embodiment of the invention, the log solving processing is carried out on the time length value of the v-th video watched by the u-th user in the set time length and the first average time length value and the second average time length value, so that the influence of the long-tail distribution characteristic of the time length value on the preference value of the explicit conversion calculation user can be reduced, namely the influence caused by too large magnitude difference of the time length value is reduced, and the accuracy of the explicit conversion is improved.
A user preference value and the first average duration valueIn inverse proportion. This is because different users have different user behavior habits, and thus the first average duration values for watching the video are greatly different. Under the condition that the first average duration value of the video watched by the user is equal, the user with smaller value can better express the preference of the user to the video compared with the user with larger first average duration value, so that the user preference value is equal to the first average duration valueIn inverse proportion.
A user preference value and the second average duration valueIn inverse proportion. This is because different videos have different factors such as material, style, quality, etc., which cause a large difference in the magnitude of the duration value of the video watched by the user, for example, the duration of watching a movie is significantly longer than the duration of watching a music video, so that the preference of the user for music programs is significantly higher than the preference for movies when the duration values of the videos watched by the user are equal, and thus it can be seen that the user preference value and the first average duration value are differentIn inverse proportion.
User preference value and user video watching time length value tu,vAnd is proportional because the length of time a user watches a video may reflect the user's preference for that type of video.
Based on the above analysis, i.e. the user preference value and the first average duration valueInversely proportional to said second mean time length valueInversely proportional to the value of the time length t for the user to watch the videou,vIn proportion, the following formula can be obtained to calculate the preference value of the u user for the v video:
wherein r isu,vIndicates the preference value, t, of the u-th user for the v-th videou,vA value representing a duration of time during which the uth user viewed the vth video within the set period of time,representing the value of the first average duration,representing the second average duration value.
Wherein, according to the formula (1) in the step 202 and the formula (2) in the step 203, the formula (3) can be expanded as follows:
wherein r isu,vIndicates the preference value, t, of the u-th user for the v-th videou,vA value representing a duration of time during which the uth user viewed the vth video within the set period of time,representing the value of the first average duration,represents the second average duration value, | Uv| represents the number of all users watching the vth video within the set time length, | VuL represents the number of videos that the u-th user viewed within the set length of time,means for accumulating the time length values of all videos watched by the u-th user within the set time length,and the time length value of each user watching the v-th video in the set of users watching the v-th video in the set time length is accumulated.
Specifically, the preference value r of the u-th user to the v-th videou,vReflecting the preference degree of the u-th user to the v-th video, and if the preference value is high, reflecting the high preference degree of the u-th user to the v-th video; and if the preference value is low, reflecting that the preference degree of the u-th user to the v-th video is low.
Preferably, the calculated preference values of the u-th user for all videos in the video set may be normalized to be converted into the interval [1,10 ].
The above process is described by taking the calculation of the preference value of the u-th user for the v-th video as an example, and it should be understood that the preference value of the u-th user for each video in the video set within the set time length can be calculated according to the above process by the u-th user. It should also be appreciated that for each user who has viewed the video for the set length of time, the preference value for each user for each video in the video set may be calculated in the manner described above.
Specifically, a user preference record is obtained according to the video preference value of the user obtained through calculation, wherein the user preference record comprises a user identifier, a video identifier and a preference value.
In particular implementations, the resulting user preference record may be stored by the user preference determination system in a user preference record database, which may be a stand-alone or integrated database in the video platform.
Preferably, the user preference determining system for implementing the method provided by the embodiment of the present invention may periodically obtain the video record according to a preset period, calculate the preference value of the user for the video, obtain the user preference record, and update the user preference record database.
Fig. 3 is a schematic flowchart of a video recommendation method implemented based on a user preference record obtained by a user preference determination method according to an embodiment of the present invention, where the flowchart may be implemented by a video recommendation system, and the flowchart includes the following steps:
step 301: and acquiring a user identifier.
Specifically, in some embodiments, the manner of obtaining the user identifier may include receiving a video access request sent by the user equipment, and obtaining the user identifier carried in the video access request. The video access request can be a request for a user to log in a video system, or a video query request, or a video on demand request, etc.
In other embodiments, the video recommendation system generally maintains information about users currently logged into the system, and accordingly, the system may obtain the user identification based on the user information it maintains.
Step 302: and inquiring the user preference record according to the user identification.
In specific implementation, the video recommendation system queries and acquires a user preference record corresponding to the user identifier from the user preference record database according to the user identifier.
Step 303: and sending video recommendation information to the user equipment according to the query result.
In specific implementation, the video recommendation system can generate video recommendation information containing video information corresponding to the video identification according to the video identification in the user preference record with a high user preference value through the inquired user preference record, and send the video recommendation information to the user equipment according to the user identification.
Preferably, the video recommendation system may also obtain the user preference record by periodically accessing the user preference record database, obtain the user identifier according to the user preference record, and actively send the video recommendation information to the user.
The video recommendation information can be presented on the user equipment through texts, pictures, videos and the like.
As can be seen from the above description, in the embodiment of the present invention, user preference determination is implemented based on video recording, that is, by obtaining video recording within a set time length, where the video recording includes user implicit feedback data that does not directly indicate a user's preference degree, such as a time length value of a video watched by a user, and performing explicit conversion on implicit feedback of these users through calculation, that is, according to the obtained video recording, a time length value of a vth video watched by a vth user within the set time length, a first average time length value of the vth video watched by the vth user within the set time length, and a second average time length value of the vth video watched by all users watching the vth video within the set time length, which are obtained through calculation, a preference value of the user for the video is obtained, so as to obtain a user preference record. Based on the user preference record obtained by the user preference determining method, video recommendation is realized, namely, the user preference record is inquired according to the user identification by obtaining the user identification, and video recommendation information is sent to the user equipment according to the inquiry result. It can be seen that in the embodiment of the present invention, only the video record needs to be acquired, and the operation of the user equipment is not needed, and the record can be directly recorded by the video platform, so that no additional network resource needs to be occupied, and the user experience is not affected. In addition, the embodiment of the invention performs logarithm processing on the time length value contained in the video record and the time length value obtained in the calculation, reduces the influence of the long tail distribution characteristic of the actual time length value on the explicit conversion and the calculation of the user preference value, reduces the influence caused by overlarge magnitude difference of the time length value, improves the accuracy of implicit feedback explicit conversion, and further improves the accuracy of user preference determination and the accuracy of video recommendation.
In order to more clearly understand the application of the user preference determining method and the video recommendation method implemented based on the user preference record in a specific actual scene provided by the embodiment of the present invention, the flows shown in fig. 2 and fig. 3 are described by taking a scene applied in a specific video platform as an example. Specifically, the user preference determining system and the video recommending system provided by the embodiments of the present invention may be integrated in a large-scale video platform, where the video platform has a video recording database and a user preference recording database, a user joins a user group of the video platform through registration or the like, and when the video platform detects that a behavior of watching a video in a video resource library of the video platform exists in a user group of the user, the video platform records the watching behavior to obtain a video recording, where the video recording includes a user identifier of the user, a video identifier of the video, a time when the user starts watching the video, an expiration time when the user watches the video, a time length when the user watches the video, and other information. The video record database stores the video record in real time. A video record database may store all video records of the user since registering with the video platform. The video platform can preset the set time length for the user preference determination system to acquire the video records in the video record database, and the time length can also be changed or set by submitting an external instruction through an administrator. The user preference determining system acquires video records in a video record database according to the set duration, calculates to obtain a user preference value, generates a user preference record containing information such as user identification, the user preference value, the video identification and the like, and stores the user preference record in a user preference record database. When a video recommendation system receives a video request message sent by user equipment, according to a user identifier contained in the message, inquiring and acquiring a user preference record of the user from a user preference record database, selecting the user preference record with a high user preference value, generating video recommendation information containing the user identifier according to the video identifier contained in the user preference record database, and performing video recommendation on the user equipment by text, thumbnail pictures or a video preview mode.
Further, in order to more clearly understand the technical effects that can be achieved by the video recommendation method implemented by the video-recording-based user preference determination method and the user preference recording obtained by the user preference determination method according to the embodiment of the present invention, based on the flow shown in fig. 2, fig. 4 to 7 show the calculation effects of applying the video-recording-based user preference determination method according to the embodiment of the present invention to an actual scene.
The actual scene is specifically 22404 video records of 4132 videos obtained by 7731 users in a month within a set time length of the month. Wherein, fig. 4 shows the video viewing time length value distribution of users in the actual application scene in the month, it can be seen that the time length values (vertical axis) of 7731 users (horizontal axis) in the actual scene for viewing 4132 videos present a long tail distribution, at the head of the curve, the time length value for viewing the video is very large, then the curve drops sharply, but does not drop to zero value at the tail of the curve, but rather approaches the horizontal axis very slowly, so the viewing time length magnitude is very different; fig. 5 shows the first average duration value (vertical axis) distribution in the actual application scene, and likewise, it can be seen that the first average duration values of 7731 users (horizontal axis) watching videos in the actual scene also present a long tail distribution; fig. 6 shows a second average temporal value (vertical axis) distribution in the actual application scene, and likewise, it can be seen that the second average temporal values of 4132 videos (horizontal axis) in the actual scene also exhibit a long-tailed distribution; finally, fig. 7 shows the result of applying the video recording-based user preference determination method provided by the embodiment of the present invention to the actual scene, calculating the obtained user preference value, and further converting the normalized user preference value display into the interval [1,10 ]. It can be seen that in the histogram shown in fig. 7, the normalized user preference values are concentrated in the interval [6,8], which is consistent with the distribution of user preference values obtained by explicit scoring of users in the public data set in the actual application scenario. The average value obtained by calculating the average value of the normalized user preference value result obtained by the method provided by the embodiment of the invention is about 6.2, which indicates that the user prefers to give a video positive evaluation, and the result is consistent with the conclusion obtained by explicitly scoring the public data set. Therefore, the method for determining the user preference based on the video recording provided by the embodiment of the invention can determine the user preference based on the video recording, and can determine the user preference more accurately in practical application.
Based on the same technical concept, the embodiment of the invention also provides a user preference determining system based on video recording, and the system can realize the embodiment of the user preference determining method based on video recording.
Fig. 8 is a schematic structural diagram of a user preference determining system according to an embodiment of the present invention, where the system may include: an obtaining module 801, a first calculating module 802, and a second calculating module 803, wherein:
an obtaining module 801, configured to obtain a video record within a set time length to obtain a video set and a user set, where a video in the video set is a video that has been viewed within the set time length, a user in the user set is a user that has viewed the video within the set time length, the video set includes M videos, the user set includes N users, and M, N are integers greater than or equal to 1;
a first calculating module 802, configured to calculate, according to the obtained video record, a first average duration value of the u-th user watching a video within the set time duration for the u-th user in the user set, where u is greater than or equal to 1 and is less than or equal to N; the video acquisition unit is used for acquiring a video set of a user, and calculating a second average time length value of all users watching the v-th video in the set time length aiming at the v-th video in the video set according to the acquired video record, wherein v is more than or equal to 1 and less than or equal to M;
a second calculating module 803, configured to calculate a preference value of the u user for the v-th video according to the time length value of the u user watching the v-th video within the set time length, the first average time length value, and the second average time length value, so as to obtain a user preference record, where the user preference record includes a user identifier, a video identifier, and a preference value.
The video recording refers to a recording formed by a user watching a video, and can record the content of the video watched by the user, such as an identifier, watching time (which may include a watching start time and a watching end time), watching duration and the like. Specifically, the video record may include information such as a user identifier, a video identifier, a timestamp, and a time length value for the user to watch the video. The video recording may be stored in a video recording database.
Specifically, the set time length in the obtaining module 801 may be pre-configured in the user preference determining system, and the embodiment of the present invention also allows setting the time length or modifying the pre-configured time length by manually submitting an instruction. The set time period may be a period of one week or one month. When a video viewing record within a set period of time needs to be acquired, the user preference determination system may call the video record within the set period of time from the video record database according to the viewing time in the video record database, including information of the viewing start time and the viewing end time.
Specifically, the video set obtained by the obtaining module 801 may be a set of videos that have been viewed within a set time length, or may be a video set obtained by obtaining a video record corresponding to a video type within a set time length according to the set video type. The video set comprises M videos, which can be arranged from a start position 1 to an end position M in the video set according to the video identifiers in the video record.
Specifically, the user set obtained by the obtaining module 801 is a set of users who have viewed videos within a set time length, or may be a user set obtained by obtaining video records corresponding to the user type within the set time length according to the set user type. The user set comprises N users, and the users can be arranged from a starting position 1 to an ending position N in the user set according to the user identification in the video recording, wherein M, N are integers which are more than or equal to 1. .
Specifically, the first computing module 802 is configured to: the first average duration value is calculated according to equation (1).
Specifically, the first computing module 802 is further configured to: the second average duration value is calculated according to equation (2).
Specifically, the second calculation module 803 is used for: respectively carrying out logarithm solving processing on the duration value of the vth video watched by the u user in the set time length, the first average duration value and the second average duration value; and calculating the preference value of the u user to the v video according to the time length value of the v video watched by the u user in the set time length and the result of logarithmic processing of the first average time length value and the second average time length value.
Wherein, the user preference value and the time length value t of the video watched by the useru,vIs proportional to the first average time length valueSecond average time length valueIn inverse proportion.
Specifically, the second calculation module 803 is further configured to: and (4) calculating the preference value of the u user to the v video according to the formula (3).
Further, the second calculating module 803 is further configured to calculate the preference value of the u-th user for the v-th video according to formula (1), formula (2) and formula (3) by using formula (4).
Specifically, the preference value r of the u-th user to the v-th videou,vReflecting the preference degree of the u-th user to the v-th video, and if the preference value is high, reflecting the high preference degree of the u-th user to the v-th video; and if the preference value is low, reflecting that the preference degree of the u-th user to the v-th video is low.
Preferably, the second calculating module 803 may perform a normalization operation on the calculated preference values of the u-th user for all videos in the video set, so as to convert into the interval [1,10 ].
Specifically, a user preference record is obtained according to the video preference value of the user obtained through calculation, wherein the user preference record comprises a user identifier, a video identifier and a preference value.
In particular implementations, the resulting user preference record may be stored by the user preference determination system in a user preference record database, which may be a stand-alone or integrated database in the video platform.
Based on the same technical concept, the embodiment of the invention also provides a video recommendation system realized based on the user preference record obtained by the user preference determination method.
Fig. 9 shows a schematic structural diagram of a video recommendation system according to an embodiment of the present invention, where the system may include: an obtaining module 901, a query module 902, and a recommending module 903, wherein:
an obtaining module 901, configured to obtain a user identifier;
a query module 902, configured to query the user preference record according to the user identifier;
and the recommending module 903 is used for sending video recommending information to the user equipment according to the query result.
Specifically, in some embodiments, the manner of acquiring the user identifier by the acquiring module 901 may include receiving a video access request sent by the user equipment, and acquiring the user identifier carried in the video access request. The video access request can be a request for a user to log in a video system, or a video query request, or a video on demand request, etc.
In other embodiments, the video recommendation system generally maintains information about users currently logged into the system, and accordingly, the system may obtain the user identification based on the user information it maintains.
In specific implementation, the video recommendation module 903 may generate video recommendation information including video information corresponding to a video identifier according to the user preference record queried by the query module 902 and according to the video identifier in the user preference record with a high user preference value. And sending video recommendation information to user equipment corresponding to the user identification. The video recommendation information can be presented on the user equipment through texts, pictures, videos and the like.
In summary, in the above embodiments of the present invention, the determination of the user preference is implemented based on video recording, that is, a video set and a user set are obtained by obtaining video recording within a set time length, where videos in the video set are videos that have been viewed within the set time length, users in the user set are users that have viewed videos within the set time length, a first average time length value of the u-th user viewing videos within the set time length and a second average time length value of all users viewing the v-th videos within the set time length are calculated according to the obtained video recording, and a preference value of the u-th user for the v-th video is calculated according to a time length value of the u-th user viewing the v-th video within the set time length, and the first average time length value and the second average time length value, and obtaining a user preference record, and realizing video recommendation based on the user preference record obtained by the user preference determination method. It can be seen that in the method provided by the embodiment of the present invention, only the obtained video record is needed, and the operation of the user equipment is not needed, and the video record can be directly recorded by the video platform, and does not occupy additional network resources and does not affect the experience of the user. Meanwhile, logarithm processing is carried out on the used time length value in the embodiment of the invention, so that the influence of the long tail distribution characteristic of the time length value on the explicit conversion is reduced, namely the influence caused by too large magnitude difference of the time length value is reduced, the accuracy of the user preference determination is improved, and the video recommendation can be carried out more accurately. Therefore, the method and the system for determining the user preference based on the video recording and the method and the system for recommending the video based on the user preference recording obtained by the method for determining the user preference realize the determination of the user preference to the video and the video recommendation by using the resources which are easy to obtain in the network.
The present invention is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each flow and/or block of the flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
While preferred embodiments of the present invention have been described, additional variations and modifications in those embodiments may occur to those skilled in the art once they learn of the basic inventive concepts. Therefore, it is intended that the appended claims be interpreted as including preferred embodiments and all such alterations and modifications as fall within the scope of the invention.
It will be apparent to those skilled in the art that various changes and modifications may be made in the present invention without departing from the spirit and scope of the invention. Thus, if such modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include such modifications and variations.
Claims (10)
1. A method for determining user preference based on video recording, comprising:
acquiring video records within a set time length to obtain a video set and a user set, wherein videos in the video set are videos watched within the set time length, users in the user set are users who have watched within the set time length, the video set comprises M videos, the user set comprises N users, and M, N are integers greater than or equal to 1;
according to the obtained video record, aiming at the u-th user in the user set, calculating a first average time length value of the u-th user watching the video within the set time length, wherein u is more than or equal to 1 and less than or equal to N;
according to the obtained video record, aiming at the v-th video in the video set, calculating a second average time length value of the v-th video watched by all users watching the v-th video within the set time length, wherein v is more than or equal to 1 and less than or equal to M;
and calculating the preference value of the u user to the v video according to the time length value of the v video watched by the u user in the set time length, the first average time length value and the second average time length value to obtain a user preference record, wherein the user preference record comprises a user identifier, a video identifier and a preference value.
2. The method of claim 1, wherein calculating, according to the obtained video record, a first average duration value for the u-th user in the user set to watch the video within the set duration comprises:
calculating a first average duration value of the video watched by the u user within the set duration according to the following formula:
wherein,represents the first average duration value, | VuL represents the number of videos watched by the u-th user within the set time length, tu,vAnd the time length value of the nth user watching the vth video in the set time length is represented.
3. The method according to claim 1, wherein calculating, according to the obtained video record, a second average duration value of the vth video watched by all users watching the vth video within the set duration for the vth video in the video set, includes:
calculating a second average duration value of the vth video watched by all users watching the vth video within the set duration according to the following formula:
wherein,represents the second average duration value, | UvL represents all watched within the set time periodNumber of users of the v-th video, tu,vAnd the time length value of the nth user watching the vth video in the set time length is represented.
4. The method of claim 1, wherein calculating the preference value of the vth user for the vth video according to the duration value of the vth video viewed by the vth user in the set duration and the first average duration value and the second average duration value comprises:
respectively carrying out logarithm solving processing on the duration value of the vth video watched by the u user in the set time length, the first average duration value and the second average duration value;
calculating the preference value of the u user to the v video according to the following formula:
wherein r isu,vIndicates the preference value, t, of the u-th user for the v-th videou,vA value representing a duration of time during which the uth user viewed the vth video within the set period of time,representing the value of the first average duration,representing the second average duration value.
5. The method of any one of claims 1 to 4, wherein obtaining a video recording for a set length of time to obtain a video set comprises:
according to a set video type, acquiring a video record corresponding to the video type within a set time length to obtain a video set; or
Acquiring video records within a set time length to obtain a user set, wherein the method comprises the following steps:
and acquiring the video record corresponding to the user type within a set time length according to the set user type to obtain a user set.
6. A video recommendation method implemented based on a user preference record derived by the method of any of claims 1 to 4, comprising:
acquiring a user identifier;
inquiring the user preference record according to the user identification;
and sending video recommendation information to the user equipment according to the query result.
7. A user preference determination system implemented based on video recording, comprising:
the acquisition module is used for acquiring video records within a set time length to obtain a video set and a user set, wherein videos in the video set are videos watched within the set time length, users in the user set are users who have watched within the set time length, the video set comprises M videos, the user set comprises N users, and M, N are integers greater than or equal to 1;
the first calculation module is used for calculating a first average time length value of the u user watching the video within the set time length according to the obtained video record, wherein u is more than or equal to 1 and less than or equal to N; the video acquisition unit is used for acquiring a video set of a user, and calculating a second average time length value of all users watching the v-th video in the set time length aiming at the v-th video in the video set according to the acquired video record, wherein v is more than or equal to 1 and less than or equal to M;
and the second calculation module is used for calculating the preference value of the u user to the v video according to the time length value of the u user watching the v video in the set time length, the first average time length value and the second average time length value to obtain a user preference record, wherein the user preference record comprises a user identifier, a video identifier and a preference value.
8. The system of claim 7, wherein the first computing module is specifically configured to: calculating the first average duration value according to the following formula:
wherein,represents the first average duration value, | VuL represents the number of videos watched by the u-th user within the set time length, tu,vA time length value representing that the u user watches the v video within the set time length; or
The first calculation module is specifically configured to: calculating the second average duration value according to the following formula:
wherein,represents the second average duration value, | UvL represents the number of all users watching the vth video in the set time length, tu,vAnd the time length value of the nth user watching the vth video in the set time length is represented.
9. The system of claim 7, wherein the second computing module is specifically configured to:
respectively carrying out logarithm solving processing on the duration value of the vth video watched by the u user in the set time length, the first average duration value and the second average duration value;
calculating the preference value of the u user to the v video according to the following formula：
Wherein r isu,vIndicates the preference value, t, of the u-th user for the v-th videou,vA value representing a duration of time during which the uth user viewed the vth video within the set period of time,representing the value of the first average duration,representing the second average duration value.
10. A video recommendation system implemented based on a user preference record derived by the system of any of claims 7 to 9, comprising:
the acquisition module is used for acquiring a user identifier;
the query module is used for querying the user preference record according to the user identification;
and the recommendation module is used for sending video recommendation information to the user equipment according to the query result.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510570624.1A CN105163142B (en) | 2015-09-09 | 2015-09-09 | A kind of user preference determines method, video recommendation method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510570624.1A CN105163142B (en) | 2015-09-09 | 2015-09-09 | A kind of user preference determines method, video recommendation method and system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN105163142A true CN105163142A (en) | 2015-12-16 |
| CN105163142B CN105163142B (en) | 2018-10-26 |
Family
ID=54803882
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510570624.1A Active CN105163142B (en) | 2015-09-09 | 2015-09-09 | A kind of user preference determines method, video recommendation method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105163142B (en) |
Cited By (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106604137A (en) * | 2016-12-29 | 2017-04-26 | Tcl集团股份有限公司 | Method and apparatus for predicting video viewing time length |
| CN107943932A (en) * | 2017-11-22 | 2018-04-20 | 广州虎牙信息科技有限公司 | Category recommends method, storage device and terminal |
| CN108134950A (en) * | 2017-12-07 | 2018-06-08 | 上海斐讯数据通信技术有限公司 | A kind of intelligent video recommends method and system |
| CN108322827A (en) * | 2018-01-31 | 2018-07-24 | 武汉斗鱼网络科技有限公司 | Method, system and the computer readable storage medium of the video preference of measure user |
| CN109388970A (en) * | 2018-10-25 | 2019-02-26 | 麒麟合盛网络技术股份有限公司 | Data processing method and device |
| CN109729392A (en) * | 2019-01-15 | 2019-05-07 | 深圳市云歌人工智能技术有限公司 | The method, apparatus and storage medium of video push |
| CN111108494A (en) * | 2017-08-17 | 2020-05-05 | 开放电视公司 | Multimedia focusing |
| CN111246254A (en) * | 2018-11-28 | 2020-06-05 | 北京字节跳动网络技术有限公司 | Video recommendation method and device, server, terminal equipment and storage medium |
| CN112825085A (en) * | 2019-11-20 | 2021-05-21 | 珠海格力电器股份有限公司 | File push optimization method, storage medium, device and terminal equipment |
| WO2023029112A1 (en) * | 2021-08-31 | 2023-03-09 | 广东艾檬电子科技有限公司 | Video recommendation method, terminal device and computer readable storage medium |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102957950A (en) * | 2012-07-23 | 2013-03-06 | 华东师范大学 | User implicit rating method for recommending video |
| US20130297412A1 (en) * | 2012-05-02 | 2013-11-07 | International Business Machines Corporation | Delivering personalized recommendations that relate to transactions on display |
| CN103686236A (en) * | 2013-11-19 | 2014-03-26 | 乐视致新电子科技（天津）有限公司 | Method and system for recommending video resource |
| CN103686237A (en) * | 2013-11-19 | 2014-03-26 | 乐视致新电子科技（天津）有限公司 | Method and system for recommending video resource |
| CN104077339A (en) * | 2013-07-09 | 2014-10-01 | 腾讯科技（深圳）有限公司 | Multimedia data user preference degree acquiring method, device and system |
-
2015
- 2015-09-09 CN CN201510570624.1A patent/CN105163142B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20130297412A1 (en) * | 2012-05-02 | 2013-11-07 | International Business Machines Corporation | Delivering personalized recommendations that relate to transactions on display |
| CN102957950A (en) * | 2012-07-23 | 2013-03-06 | 华东师范大学 | User implicit rating method for recommending video |
| CN104077339A (en) * | 2013-07-09 | 2014-10-01 | 腾讯科技（深圳）有限公司 | Multimedia data user preference degree acquiring method, device and system |
| CN103686236A (en) * | 2013-11-19 | 2014-03-26 | 乐视致新电子科技（天津）有限公司 | Method and system for recommending video resource |
| CN103686237A (en) * | 2013-11-19 | 2014-03-26 | 乐视致新电子科技（天津）有限公司 | Method and system for recommending video resource |
Non-Patent Citations (1)
| Title |
|---|
| 梁昌勇等: "电子商务推荐系统中群体用户推荐问题研究", 《中国管理科学》 * |
Cited By (15)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106604137A (en) * | 2016-12-29 | 2017-04-26 | Tcl集团股份有限公司 | Method and apparatus for predicting video viewing time length |
| CN111108494B (en) * | 2017-08-17 | 2023-11-28 | 开放电视公司 | Multimedia focusing |
| CN111108494A (en) * | 2017-08-17 | 2020-05-05 | 开放电视公司 | Multimedia focusing |
| CN107943932B (en) * | 2017-11-22 | 2020-06-09 | 广州虎牙信息科技有限公司 | Item recommendation method, storage device and terminal |
| CN107943932A (en) * | 2017-11-22 | 2018-04-20 | 广州虎牙信息科技有限公司 | Category recommends method, storage device and terminal |
| CN108134950A (en) * | 2017-12-07 | 2018-06-08 | 上海斐讯数据通信技术有限公司 | A kind of intelligent video recommends method and system |
| CN108134950B (en) * | 2017-12-07 | 2022-05-06 | 广州锐竞信息科技有限责任公司 | Intelligent video recommendation method and system |
| CN108322827A (en) * | 2018-01-31 | 2018-07-24 | 武汉斗鱼网络科技有限公司 | Method, system and the computer readable storage medium of the video preference of measure user |
| CN108322827B (en) * | 2018-01-31 | 2021-02-02 | 武汉斗鱼网络科技有限公司 | Method, system, and computer-readable storage medium for measuring video preferences of a user |
| CN109388970B (en) * | 2018-10-25 | 2020-08-21 | 麒麟合盛网络技术股份有限公司 | Data processing method and device |
| CN109388970A (en) * | 2018-10-25 | 2019-02-26 | 麒麟合盛网络技术股份有限公司 | Data processing method and device |
| CN111246254A (en) * | 2018-11-28 | 2020-06-05 | 北京字节跳动网络技术有限公司 | Video recommendation method and device, server, terminal equipment and storage medium |
| CN109729392A (en) * | 2019-01-15 | 2019-05-07 | 深圳市云歌人工智能技术有限公司 | The method, apparatus and storage medium of video push |
| CN112825085A (en) * | 2019-11-20 | 2021-05-21 | 珠海格力电器股份有限公司 | File push optimization method, storage medium, device and terminal equipment |
| WO2023029112A1 (en) * | 2021-08-31 | 2023-03-09 | 广东艾檬电子科技有限公司 | Video recommendation method, terminal device and computer readable storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN105163142B (en) | 2018-10-26 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105163142B (en) | A kind of user preference determines method, video recommendation method and system | |
| US9948965B2 (en) | Manifest re-assembler for a streaming video channel | |
| CN105516821B (en) | The method and device of barrage screening | |
| CN108900923B (en) | Method and device for recommending live broadcast template | |
| CN108875022B (en) | Video recommendation method and device | |
| CN103686237B (en) | Recommend the method and system of video resource | |
| CN106326391B (en) | Multimedia resource recommendation method and device | |
| WO2016054916A1 (en) | Video content recommending and evaluating methods and devices | |
| US20170188102A1 (en) | Method and electronic device for video content recommendation | |
| KR20140088187A (en) | User viewing data collection for generating media viewing achievements | |
| CN104584571A (en) | Generating a sequence of audio fingerprints at a set top box | |
| CN103686375A (en) | Video sharing method and device | |
| CN103997662A (en) | Program pushing method and system | |
| CN107592572B (en) | Video recommendation method, device and equipment | |
| CN109462765B (en) | Method and device for issuing and displaying recommended page | |
| CN105871585A (en) | Terminal association method and device | |
| Xu et al. | Catch-up TV recommendations: show old favourites and find new ones | |
| CN108600780B (en) | Method for pushing information, electronic device and computer readable medium | |
| CN109348261A (en) | Data statistical approach, device and electronic equipment in a kind of live streaming | |
| CN106534984B (en) | Television program pushing method and device | |
| CN106303591B (en) | A kind of video recommendation method and device | |
| CN112104910A (en) | Video searching method, device and system | |
| KR101779975B1 (en) | System for providing additional service of VOD content using SNS message and method for providing additional service using the same | |
| CN109327736B (en) | Program recommendation method based on program viewing expression and user information | |
| CN116208579A (en) | Information pushing method, device, equipment and storage medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C41 | Transfer of patent application or patent right or utility model | ||
| TA01 | Transfer of patent application right |
Effective date of registration: 20170109 Address after: 266000 Shandong Province, Qingdao city Laoshan District Songling Road No. 399 Applicant after: Poly Polytron Technologies Inc Address before: 266061 Laoshan, Qingdao province Hongkong District No. East Road, room 248, room 131 Applicant before: Hisense Media Networks Co., Ltd. |
|
| GR01 | Patent grant | ||
| GR01 | Patent grant |