CN114168791A - Video recommendation method and device, electronic equipment and storage medium - Google Patents

Video recommendation method and device, electronic equipment and storage medium Download PDF

Info

Publication number
CN114168791A
CN114168791A CN202111406487.XA CN202111406487A CN114168791A CN 114168791 A CN114168791 A CN 114168791A CN 202111406487 A CN202111406487 A CN 202111406487A CN 114168791 A CN114168791 A CN 114168791A
Authority
CN
China
Prior art keywords
weight
user
interest
determining
users
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Pending
Application number
CN202111406487.XA
Other languages
Chinese (zh)
Inventor
石奕
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Zhuo Erzhi Lian Wuhan Research Institute Co Ltd
Original Assignee
Zhuo Erzhi Lian Wuhan Research Institute Co Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Zhuo Erzhi Lian Wuhan Research Institute Co Ltd filed Critical Zhuo Erzhi Lian Wuhan Research Institute Co Ltd
Priority to CN202111406487.XA priority Critical patent/CN114168791A/en
Publication of CN114168791A publication Critical patent/CN114168791A/en
Pending legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/70—Information retrieval; Database structures therefor; File system structures therefor of video data
    • G06F16/73—Querying
    • G06F16/735—Filtering based on additional data, e.g. user or group profiles
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/70—Information retrieval; Database structures therefor; File system structures therefor of video data
    • G06F16/78—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F40/00—Handling natural language data
    • G06F40/20—Natural language analysis
    • G06F40/205—Parsing
    • G06F40/216—Parsing using statistical methods

Abstract

The embodiment of the invention discloses a video recommendation method and device, electronic equipment and a storage medium. The method comprises the following steps: obtaining a plurality of first-level interest tags influencing the video interest and hobbies of a user and a plurality of second-level interest tags corresponding to each first-level interest tag; the secondary interest label represents the attribute value of the primary interest label; determining a first weight of a primary interest tag corresponding to each user in a plurality of users and a second weight and a third weight of a secondary interest tag corresponding to the primary interest tag according to videos historically browsed by the users; determining a fourth weight for said each user based on said first weight, said second weight and said third weight; the fourth weight represents the degree of the each user's preference for the video; matching preset label information of different types of users according to the fourth weight to obtain target label information of a target type; and recommending videos based on the target label information.

Description

Video recommendation method and device, electronic equipment and storage medium
Technical Field
The invention relates to the field of big data processing, in particular to a video recommendation method, a video recommendation device, electronic equipment and a storage medium.
Background
Most of current researches only consider the video interest preference criteria based on the historical data of the users to design recommendation algorithms when recommending the users, and omit the information that the common interest preference of the group layers where the users are located facing certain contents is also a component of the video interest preference of the users. In addition, the recommendation lists of all users have the same construction strategy, namely the recommendation list content of each user is only sorted according to the comprehensive similarity of the videos, the types of the videos in the recommendation list are not distinguished, and the proportion of the number of the videos of each type in the recommendation list is not required. However, no effective solution is available for this problem.
Disclosure of Invention
In view of this, embodiments of the present invention are intended to provide a video recommendation method, apparatus, electronic device, and storage medium.
The technical embodiment of the invention is realized as follows:
the embodiment of the invention provides a video recommendation method, which comprises the following steps:
obtaining a plurality of first-level interest tags influencing the video interest and hobbies of a user and a plurality of second-level interest tags corresponding to each first-level interest tag; the secondary interest label represents the attribute value of the primary interest label;
determining a first weight of a primary interest tag corresponding to each user in a plurality of users and a second weight and a third weight of a secondary interest tag corresponding to the primary interest tag according to videos historically browsed by the users; the first weight represents the degree of importance of each user on the primary interest tag; the second weight represents the timeliness degree corresponding to the time adopted by each user for the secondary interest tag; the third weight characterizes the interest degree of each user in the secondary interest tag;
determining a fourth weight for said each user based on said first weight, said second weight and said third weight; the fourth weight represents the degree of the each user's preference for the video;
matching preset label information of different types of users according to the fourth weight to obtain target label information of a target type;
and recommending videos based on the target label information.
In the above solution, the plurality of primary interest tags at least includes: each video corresponds to a director, actors, genre, drama, distribution area, language, winning award, and source of script.
In the above scheme, the determining a first weight of each user in the plurality of users corresponding to a primary interest tag according to videos historically browsed by the plurality of users includes:
determining the number and the maximum word frequency of the primary interest labels corresponding to the videos selected by each user according to the videos historically browsed by the users;
and determining the first weight according to the number and the maximum word frequency.
In the foregoing solution, the determining, according to videos historically browsed by multiple users, a second weight of each user in the multiple users under the primary interest tag corresponding to the secondary interest tag includes:
obtaining a first time when each user selects the secondary interest tag corresponding to the video under the primary interest tag and a second time when the video is watched for the first time according to videos historically browsed by a plurality of users;
determining the second weight according to the first time and the second time.
In the foregoing solution, the determining, according to videos historically browsed by multiple users, a third weight of each user in the multiple users under the primary interest tag corresponding to the secondary interest tag includes:
obtaining the times that each user selects the secondary interest tag corresponding to the video under the primary interest tag to be adopted and the total number of watched videos according to videos historically browsed by a plurality of users;
and determining the third weight according to the times and the total number.
In the foregoing solution, the determining the fourth weight of each user based on the first weight, the second weight and the third weight includes:
obtaining an adjusting factor for adjusting the first weight and a balance factor for weighing the second weight and the third weight distribution;
determining a composite weight for each secondary interest tag based on the second weight, the third weight, and the balance factor;
determining the fourth weight according to the first weight, the adjustment factor, the second weight, the third weight, and the balance factor.
The embodiment of the invention provides a video recommendation device, which comprises: the device comprises an obtaining unit, a first determining unit, a second determining unit, a matching unit and a recommending unit, wherein:
the obtaining unit is used for obtaining a plurality of first-level interest tags influencing the video interest and hobbies of the user and a plurality of second-level interest tags corresponding to each first-level interest tag; the secondary interest label represents the attribute value of the primary interest label;
the first determining unit is used for determining a first weight corresponding to a primary interest tag of each user in a plurality of users and a second weight and a third weight corresponding to a secondary interest tag under the primary interest tag according to videos historically browsed by the users; the first weight represents the degree of importance of each user on the primary interest tag; the second weight represents the timeliness degree corresponding to the time adopted by each user for the secondary interest tag; the third weight characterizes the interest degree of each user in the secondary interest tag;
the second determining unit is configured to determine a fourth weight of each user based on the first weight, the second weight, and the third weight; the fourth weight represents the degree of the each user's preference for the video;
the matching unit is used for matching preset label information of different types of users according to the fourth weight to obtain target label information of a target type;
and the recommending unit is used for recommending videos based on the target label information.
In the above solution, the plurality of primary interest tags at least includes: each video corresponds to a director, actors, genre, drama, distribution area, language, winning award, and source of script.
In the above scheme, the first determining unit is further configured to determine, according to videos historically browsed by multiple users, the number and the maximum word frequency of the primary interest tags corresponding to the videos selected by each user; and determining the first weight according to the number and the maximum word frequency.
In the above scheme, the first determining unit is further configured to obtain, according to videos historically browsed by multiple users, a first time when each user selects that the secondary interest tag corresponding to the video under the primary interest tag is adopted and a second time when the video is initially viewed; determining the second weight according to the first time and the second time.
In the above scheme, the first determining unit is further configured to obtain, according to videos historically browsed by multiple users, the number of times that each user selects that the secondary interest tag corresponding to the video under the primary interest tag is adopted and the total number of videos to be watched; and determining the third weight according to the times and the total number.
In the above scheme, the second determining unit is further configured to obtain an adjustment factor for adjusting the first weight and a balance factor for balancing the second weight and the third weight; determining a composite weight for each secondary interest tag based on the second weight, the third weight, and the balance factor; determining the fourth weight according to the first weight, the adjustment factor, the second weight, the third weight, and the balance factor.
In the above solution, the second determining unit is further configured to determine a fifth weight of each secondary interest tag based on the second weight, the third weight and the balance factor; determining the fourth weight according to the first weight, the adjustment factor, and the fifth weight.
An embodiment of the present invention provides an electronic device, including: a processor and a memory for storing a computer program operable on the processor, wherein the processor is operable to perform any of the steps of the method described above when executing the computer program.
Embodiments of the present invention provide a computer-readable storage medium, on which a computer program is stored, which, when being executed by a processor, implements any of the steps of the above-mentioned method.
The embodiment of the invention provides a video recommendation method, a video recommendation device, electronic equipment and a storage medium, wherein a plurality of primary interest tags influencing video interests of a user and a plurality of secondary interest tags corresponding to each primary interest tag are obtained; the secondary interest label represents the attribute value of the primary interest label; determining a first weight of a primary interest tag corresponding to each user in a plurality of users and a second weight and a third weight of a secondary interest tag corresponding to the primary interest tag according to videos historically browsed by the users; the first weight represents the degree of importance of each user on the primary interest tag; the second weight represents the timeliness degree corresponding to the time adopted by each user for the secondary interest tag; the third weight characterizes the interest degree of each user in the secondary interest tag; determining a fourth weight for said each user based on said first weight, said second weight and said third weight; the fourth weight represents the degree of the each user's preference for the video; matching preset label information of different types of users according to the fourth weight to obtain target label information of a target type; the video recommendation is performed based on the target tag information, the first-level weight and the second-level weight of the user interest tag are considered, the feedback update is timely performed, and the video recommendation effect for the user is better and more accurate.
Drawings
FIG. 1 is a schematic view illustrating a flow chart of a video recommendation method according to an embodiment of the present invention;
FIG. 2 is a schematic diagram of a configuration of a video recommendation apparatus according to an embodiment of the present invention;
fig. 3 is a schematic diagram of a hardware entity structure of an electronic device according to an embodiment of the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the embodiments of the present invention clearer, the following describes specific technical solutions of the present invention in further detail with reference to the accompanying drawings in the embodiments of the present invention. The following examples are intended to illustrate the invention but are not intended to limit the scope of the invention.
The present embodiment provides a video recommendation method, and fig. 1 is a schematic flow chart of an implementation of the video recommendation method according to the embodiment of the present invention, as shown in fig. 1, the method includes:
step S101: obtaining a plurality of first-level interest tags influencing the video interest and hobbies of a user and a plurality of second-level interest tags corresponding to each first-level interest tag; the secondary interest label represents the attribute value of the primary interest label.
It should be noted that the video recommendation method may be a video recommendation method based on a user interest representation.
The primary interest tag can represent the attribute of the video; the plurality of primary interest tags may be determined according to actual conditions, and are not limited herein. As an example, the plurality of primary interest tags may include at least a director, actors, genre, drama, distribution area, language, winning cases, and script source for each video.
The secondary interest label represents the attribute value of the primary interest label; the plurality of secondary interest tags may be determined according to actual conditions, and are not limited herein. As an example, in the case that the primary interest tag is a director, the secondary interest tag corresponding to the primary interest tag may be a zhangbi, chenkege, and the like.
Step S102: determining a first weight of a primary interest tag corresponding to each user in a plurality of users and a second weight and a third weight of a secondary interest tag corresponding to the primary interest tag according to videos historically browsed by the users; the first weight represents the degree of importance of each user on the primary interest tag; the second weight represents the timeliness degree corresponding to the time adopted by each user for the secondary interest tag; the third weight characterizes a degree of interest of the each user in the secondary interest tag.
Here, the first weight represents the degree of importance of each user to the primary interest tag, mainly considering that when each user selects a video, the degree of importance of each user to each primary interest tag of the video is different. Such as: some users may place more emphasis on the director tab and some users may place more emphasis on the actor tab.
Determining a first weight of a primary interest tag corresponding to each user in the plurality of users according to videos historically browsed by the plurality of users can be understood as calculating the weights of the primary interest tags of the plurality of users according to the videos historically browsed by the plurality of users; wherein, the plurality of users can be understood as group users; the first weight may be referred to as a primary interest tag weight; the first weight may be denoted as w1i(ii) a As an example, determining a first weight of a primary interest tag corresponding to each of a plurality of users according to videos historically browsed by the plurality of users may determine, for the videos historically browsed by the plurality of users, the number and the maximum word frequency of the primary interest tags corresponding to the videos selected by each user; and determining the first weight according to the number and the maximum word frequency.
The second weight represents the timeliness degree corresponding to the time adopted by each user for the secondary interest tag, and the timeliness and the dynamics of the user interest which can dynamically change along with the time are mainly considered, so that the timeliness weight of the secondary interest tag is included when the user video interest preference is calculated. The closer the record in the user historical browsing record to the current recommendation time is, the more real the user video interest is conveyed, and correspondingly, the higher the weight of the secondary interest label contained in the record on the constructed user video interest image is, the stronger the timeliness is.
Said is in accordance withDetermining a second weight of each user in the plurality of users corresponding to the secondary interest tags under the primary interest tags by videos historically browsed by the users; wherein, the second weight may also be referred to as an age indicator weight; the second weight may also be denoted as w2i(ii) a The determining of the second weight of each user corresponding to the secondary interest tag under the primary interest tag according to the videos browsed by the plurality of users in the history can be understood as calculating the age index weight of the secondary interest tag of the user according to the videos browsed by the plurality of users in the history. As an example, the determining, according to videos historically browsed by multiple users, the second weight of each user in the multiple users corresponding to the primary interest tag under the primary interest tag may obtain, for videos historically browsed by multiple users, a first time at which each user selects the secondary interest tag corresponding to the video under the primary interest tag to be adopted and a second time at which the video is initially watched; determining the second weight according to the first time and the second time.
The third weight represents the interest degree of each user in the secondary interest tags, and mainly considers the general situation, if the secondary interest tags are adopted by the users more times, the interest value of the users in the tags is higher, and accordingly the videos containing the tags are interested. Therefore, the embodiment uses the word frequency as a basic index for measuring the weight of the secondary interest tag words.
Determining a third weight of each user in the plurality of users corresponding to the secondary interest tags under the primary interest tags according to videos historically browsed by the users; the third weight may also be referred to as a base index weight; the third weight may be denoted as w3j(ii) a The determining of the third weight of each user corresponding to the secondary interest tag under the primary interest tag according to the videos browsed by the plurality of users in the history can be understood as calculating the basic index weight of the secondary interest tag of the user according to the videos browsed by the plurality of users in the history. As an example, the determining of the video viewed according to the plurality of user histories that each user in the plurality of users is to be under the primary interest tagThe third weight of the secondary interest tag can be the number of times that each user selects the secondary interest tag corresponding to the video under the primary interest tag to be adopted and the total number of videos to be watched, wherein the number of times is obtained according to videos historically browsed by a plurality of users; and determining the third weight according to the times and the total number.
Step S103: determining a fourth weight for said each user based on said first weight, said second weight and said third weight; the fourth weight characterizes a degree of the preference of each user for the video.
Here, the fourth weight represents the degree of the video enjoyed by each user, mainly considering that the average value of the scores of the video sets watched by each user historically is used as an index for judging whether the video is enjoyed by the user, and when the user score is greater than or equal to the average value, the video is regarded as a high-degree-of-enjoyment video, which indicates that the user likes the video very much; when the user score is lower than the average value, the video is judged to be the video with low favor degree, and the user is dissatisfied with the video. In practical applications, the fourth weight may also be referred to as an integrated weight; the fourth weight may be denoted as Wend。
Determining a fourth weight of each user based on the first weight, the second weight and the third weight may be an adjustment factor for adjusting the first weight and a balance factor for balancing the allocation of the second weight and the third weight; determining the fourth weight according to the first weight, the adjustment factor, the second weight, the third weight, and the balance factor. Wherein the balance factor may include a first balance factor and a second balance factor; the sum of the first balance factor and the second balance factor is one; the first balance factor is used for measuring the proportion of the second weight; the second balance factor is used for measuring the proportion of the third weight; the first balance factor may be denoted as θ, and the first balance factor may be denoted as 1- θ. The regulatory factor can be noted as β; the beta may be a function.
Step S104: and matching preset label information of different types of users according to the fourth weight to obtain target label information of a target type.
Here, preset tag information of different types of users is matched according to the fourth weight; the different types of users may be determined according to actual situations, and are not limited herein, and as an example, the different types of users may be teenager type users, middle-aged users, elderly users, and the like. The preset tag information includes tag weight ranges of users of the same type, for example, the preset tag information includes tag weight ranges of teenage type users, tag weight ranges of middle-aged type users, and tag weight ranges of elderly type users.
Matching preset label information of different types of users according to the fourth weight, wherein the target label information of the target type is obtained by judging which type of user's label weight range the fourth weight falls into in the preset label information; if the fourth weight falls into the label weight range of the teenager type user, acquiring target label information of the teenager type user; if the fourth weight falls into the label weight range of the middle-aged type user, obtaining target label information of the middle-aged type user; and if the fourth weight falls into the label weight range of the old-age type user, obtaining the target label information of the old-age type user.
In practical application, labeling can be performed on the video according to the fourth weight; and matching according to the label information of the video and the label information of the user class with the same requirement in the same age group to obtain matched label information.
Step S105: and recommending videos based on the target label information.
It should be noted that, the recommending a video based on the target tag information may be recommending a video corresponding to the target tag information based on the target tag information.
According to the video recommendation method provided by the embodiment of the invention, a plurality of primary interest tags influencing the video interest and hobbies of a user and a plurality of secondary interest tags corresponding to each primary interest tag are obtained; the secondary interest label represents the attribute value of the primary interest label; determining a first weight of a primary interest tag corresponding to each user in a plurality of users and a second weight and a third weight of a secondary interest tag corresponding to the primary interest tag according to videos historically browsed by the users; the first weight represents the degree of importance of each user on the primary interest tag; the second weight represents the timeliness degree corresponding to the time adopted by each user for the secondary interest tag; the third weight characterizes the interest degree of each user in the secondary interest tag; determining a fourth weight for said each user based on said first weight, said second weight and said third weight; the fourth weight represents the degree of the each user's preference for the video; matching preset label information of different types of users according to the fourth weight to obtain target label information of a target type; the video recommendation is performed based on the target tag information, the first-level weight and the second-level weight of the user interest tag are considered, the feedback update is timely performed, and the video recommendation effect for the user is better and more accurate.
In an optional embodiment of the invention, the plurality of primary interest tags comprises at least: each video corresponds to a director, actors, genre, drama, distribution area, language, winning award, and source of script.
In an optional embodiment of the present invention, the determining, according to videos historically browsed by multiple users, a first weight of each user in the multiple users corresponding to a primary interest tag includes: determining the number and the maximum word frequency of the primary interest labels corresponding to the videos selected by each user according to the videos historically browsed by the users; and determining the first weight according to the number and the maximum word frequency.
In this embodiment, it is mainly considered that when each user selects a video, the importance degree of each user to each primary interest tag of the video is different. Such as: some users may place more emphasis on the director tab and some users may place more emphasis on the actor tab.
Determining the number and the maximum word frequency of the primary interest labels corresponding to the videos selected by each user according to the videos historically browsed by the users; the number of the primary interest tags can be the total number of the primary interest tags adopted by the video selected by each user; the maximum word frequency represents the primary interest label with the most times in the primary interest labels; for convenience of understanding, here, for example, a user selects a director, actors, language, winning situation, and source of a scenario corresponding to a video habit video; wherein the director takes the most number of times, for example 6 times; the number of the primary interest tags corresponding to the video selected by the user is understood to be 5; the maximum word frequency is 6 times and corresponds to a director label; i.e. the user takes more importance on the director label.
Determining the first weight according to the number and the maximum word frequency may be calculating the first weight according to the number and the maximum word frequency by a preset formula. The preset formula may be determined according to an actual situation, and is not limited herein, and as an example, the preset formula may refer to the following formula (1):
Figure BDA0003373007940000101
in the formula (1), maxi is the maximum primary interest tag word frequency in the plurality of primary interest tags, and n represents the number of the primary interest tags.
In an optional embodiment of the present invention, the determining, according to videos historically browsed by a plurality of users, a second weight of each user in the plurality of users under the primary interest tag for a corresponding secondary interest tag includes: obtaining a first time when each user selects the secondary interest tag corresponding to the video under the primary interest tag and a second time when the video is watched for the first time according to videos historically browsed by a plurality of users; determining the second weight according to the first time and the second time.
In the embodiment, the user interest is considered to be dynamically changed along with time, and the timeliness and the dynamic property are provided, so that the timeliness weight of the secondary interest tag is included when the user video interest preference is calculated. The closer the record in the user historical browsing record to the current recommendation time is, the more real the user video interest is conveyed, and correspondingly, the higher the weight of the secondary interest label contained in the record on the constructed user video interest image is, the stronger the timeliness is.
Obtaining a first time when each user selects the secondary interest tag corresponding to the video under the primary interest tag and a second time when the video is watched for the first time according to videos historically browsed by a plurality of users; the first time can be understood as the time that a certain secondary interest tag corresponding to the primary interest tag is adopted by a user each time; for ease of understanding, the certain secondary interest tag may be denoted as a secondary tag j; the first time may be denoted as timejt(ii) a The second time can be recorded as time0。
Determining the second weight according to the first time and the second time; wherein, the second weight can also be called a secondary interest tag age index weight; determining the second weight according to the first time and the second time may be determining the second weight through a preset algorithm according to the first time and the second time; the preset algorithm may be determined according to an actual situation, and is not limited herein, and as an example, the preset algorithm may refer to the following formula (2):
Figure BDA0003373007940000111
in the formula (2), timejtRepresenting the time that the secondary label j is taken by the user each time; time0Representing the time when the user first viewed the video; m represents the frequency of the secondary interest tags j, and k represents the number of all secondary interest tags.
In an optional embodiment of the present invention, the determining, according to videos historically browsed by a plurality of users, a third weight of each user in the plurality of users under the primary interest tag for a corresponding secondary interest tag includes: obtaining the times that each user selects the secondary interest tag corresponding to the video under the primary interest tag to be adopted and the total number of watched videos according to videos historically browsed by a plurality of users; and determining the third weight according to the times and the total number.
In the embodiment of the invention, general conditions are mainly considered, and if the number of times that a certain secondary interest tag is adopted by a user is more, the interest value of the user on the tag is higher, and accordingly the video containing the tag is interested. Therefore, the embodiment uses the word frequency as a basic index for measuring the weight of the secondary interest tag words.
Obtaining the times that each user selects the secondary interest tag corresponding to the video under the primary interest tag to be adopted and the total number of watched videos according to videos historically browsed by a plurality of users; wherein, the number of times can be recorded as word frequencyj(ii) a The total number of viewing videos may be denoted as Q.
Determining the third weight according to the times and the total number; wherein, the third weight may also be referred to as a secondary interest label basic index weight. In practical applications, the determining the third weight according to the number of times and the total number may be determining the third weight according to the number of times and the total number through a preset formula. The preset formula may be determined according to an actual situation, and is not limited herein, and as an example, the preset formula may refer to the following formula (3):
Figure BDA0003373007940000121
in formula (3), word frequencyjThe word frequency of the jth secondary interest tag is represented, and Q represents the total number of videos watched by the user.
In an optional embodiment of the present invention, the determining the fourth weight of each user based on the first weight, the second weight and the third weight comprises: obtaining an adjusting factor for adjusting the first weight and a balance factor for weighing the second weight and the third weight distribution; determining the fourth weight according to the first weight, the adjustment factor, the second weight, the third weight, and the balance factor.
In the embodiment of the present invention, the adjustment factor may be determined according to an actual situation, which is not limited herein, and as an example, the adjustment factor may be denoted as β; the beta may be a function. For convenience of understanding, the balance factor can be expressed by referring to the following equations (4), (5):
Figure BDA0003373007940000122
Figure BDA0003373007940000123
in the formulae (4) and (5), n1Is a constant, β is a function that adjusts the user's interest tag weight, B is a real number, and a is also a function.
The balance factor may be determined according to an actual situation, which is not limited herein, and as an example, the balance factor may include a first balance factor and a second balance factor; the sum of the first balance factor and the second balance factor is one; the first balance factor is used for measuring the proportion of the second weight; the second balance factor is used for measuring the proportion of the third weight; the first balance factor may be 0.3. The second balance factor may be 0.7; the first balance factor may be denoted as θ, and the first balance factor may be denoted as 1- θ.
Determining the fourth weight according to the first weight, the adjustment factor, the second weight, the third weight, and the balance factor may be determining the fourth weight through a preset algorithm according to the first weight, the adjustment factor, the second weight, the third weight, and the balance factor. Wherein the fourth weightMay be referred to as composite weights; the composite weight may be denoted as Wend(ii) a The preset algorithm may be determined according to actual conditions, and is not limited herein, and as an example, the preset algorithm may refer to the following formula (6):
Wend＝w1i*(θ*w2j+(1-θ)*w3j)*(1+β) (6)
in the formula (6), β is a function, and is a function for adjusting the interest tag weight of the user; theta is a balance factor used for measuring the weight distribution of the second weight and the third weight.
In an optional embodiment of the present invention, the determining the fourth weight according to the first weight, the adjustment factor, the second weight, the third weight and the balance factor comprises: determining a fifth weight for each secondary interest tag based on the second weight, the third weight, and the balance factor; determining the fourth weight according to the first weight, the adjustment factor, and the fifth weight.
In this embodiment of the present invention, a fifth weight of each secondary interest tag is determined based on the second weight, the third weight and the balance factor; the balance factor may be determined according to an actual situation, which is not limited herein, and as an example, the balance factor may include a first balance factor and a second balance factor; the sum of the first balance factor and the second balance factor is one; determining the fifth weight of each secondary interest tag based on the second weight, the third weight, and the balancing factor may be determining the fifth weight of each secondary interest tag based on the second weight, the third weight, the first balancing factor, and the second balancing factor. For convenience of understanding, the first balance factor may be denoted as θ, and the first balance factor may be denoted as 1- θ; the fifth weight may be denoted as W. Determining the fifth weight of each secondary interest tag based on the second weight, the third weight, the first balance factor, and the second balance factor may be determining the fifth weight of each secondary interest tag by a preset formula based on the second weight, the third weight, the first balance factor, and the second balance factor; the preset formula may be determined according to an actual situation, and is not limited herein, and as an example, the preset formula may refer to the following formula (7):
W＝θ*w2j+(1-θ)*w3j (7)
in equation (7), θ is a balance factor used to measure the weight distribution of the second weight and the third weight.
Determining the fourth weight according to the first weight, the adjustment factor, and the fifth weight may be determining the fourth weight according to the first weight, the adjustment factor, and the fifth weight through a preset formula; the preset formula may be determined according to an actual situation, and is not limited herein, and as an example, the preset formula may refer to the following formula (8):
Wend＝w1i*W*(1+β) (8)
in equation (8), β is a function that adjusts the user's interest tag weight.
According to the video recommendation method provided by the embodiment of the invention, the user attribute, the advertisement attribute and the relationship between the user and the advertisement are considered, and the recommended advertisement effect is better and more accurate.
In order to understand the embodiment of the present invention, the embodiment of the present invention exemplifies a video recommendation method. The method comprises the following specific steps:
firstly, selecting a user video interest preference index.
In the embodiment of the invention, the indexes of the user video interest preference model adopt six basic attribute indexes of actors, director, genre, drama editing, distribution area and language, and two indexes of video winning conditions and script sources are added.
The first-level interest tags are 8 tags of directors, actors, types, dramas, distribution areas, languages, video winning conditions and script sources; a secondary interest tag is an attribute value of its primary interest tag. For example, for a director's primary interest tag, the secondary interest tags contained therein are those of a zhangbi, chenkege, etc.
And secondly, calculating the weight of the user level interest tag.
In the embodiment of the invention, the consideration is that when each user selects the video, the attention degree of each user to each level of interest tag of the video is different. Such as: some users may place more emphasis on the director tab and some users may place more emphasis on the actor tab.
The calculation of the user's primary interest label weight can be referred to the formula (1) above.
In the formula (1), maxi is the maximum primary interest tag word frequency in the plurality of primary interest tags, and n represents the number of the primary interest tags.
And thirdly, calculating the aging index weight of the secondary interest label of the user.
In the embodiment of the invention, because the interest of the user can dynamically change along with the time and has timeliness and dynamics, the timeliness weight of the secondary interest tag is included when the interest preference of the user video is calculated. The closer the record in the user historical browsing record to the current recommendation time is, the more real the user video interest is conveyed, and correspondingly, the higher the weight of the secondary interest label contained in the record on the constructed user video interest image is, the stronger the timeliness is.
The calculation of the age index weight of the user secondary interest tag can be referred to the formula (2) above.
In the formula (2), timejtRepresenting the time that the secondary label j is taken by the user each time; time0Representing the time when the user first viewed the video; m represents the frequency of the secondary interest tags j, and k represents the number of all secondary interest tags.
And fourthly, calculating the basic index weight of the secondary interest label of the user.
In the embodiment of the invention, generally, if the number of times that a certain level of interest tags are adopted by a user is more, it indicates that the interest value of the user on the tags is higher, and accordingly, the video containing the tags is interested. Therefore, the cross-bottom book uses word frequency as a basic index for measuring the weight of the secondary interest label words.
The calculation of the user secondary interest label basic index weight can be referred to the formula (3) above.
In formula (3), word frequencyjThe word frequency of the jth secondary interest tag is represented, and Q represents the total number of videos watched by the user.
Fifth, the composite weight (Wend) of each secondary interest tag (attribute value) is calculated.
In the embodiment of the invention, the comprehensive weight of each secondary interest label is the arithmetic weighted sum of the timeliness weight w2i and the label word weight w3j, and then the sum is multiplied by the primary interest label weight w1i to which the comprehensive weight belongs, and the weight distribution of the word weight and the timeliness weight is measured by introducing the balance factor theta.
Taking the score average value of the historical watching video set of each user as an index for judging whether the user likes the video, and when the user score is larger than or equal to the average value, regarding the video as a high-likeness video, indicating that the user likes the video very much; when the user score is lower than the average value, the video is judged to be the video with low favor degree, and the user is dissatisfied with the video.
The integrated weight can be expressed by the above equations (4), (5) and (6).
In the formulae (4), (5) and (6), n1Is a constant, β is a function that adjusts the user's interest tag weight, B is a real number, and a is also a function.
Sixth, a "user representation" based on the user is generated.
In the embodiment of the invention, the video is labeled by the comprehensive label weight; matching according to the label information of the video and the label information of the user class with the same requirement in the same age group; and generating a user portrait based on the user according to the matched label information, and recommending the video.
And seventhly, feedback evaluation.
In the embodiment of the invention, the comprehensive index weight of the user is updated after a period of time, then recommendation is carried out, the feedback of the user is collected, and information updating is carried out to form a continuous circulating and updating closed loop.
The video recommendation method provided by the embodiment of the invention comprises the steps of obtaining a plurality of first-level interest tags influencing the video interest and hobbies of a user and a plurality of second-level interest tags corresponding to each first-level interest tag; the secondary interest label represents the attribute value of the primary interest label; determining a first weight of a primary interest tag corresponding to each user in a plurality of users and a second weight and a third weight of a secondary interest tag corresponding to the primary interest tag according to videos historically browsed by the users; the first weight represents the degree of importance of each user on the primary interest tag; the second weight represents the timeliness degree corresponding to the time adopted by each user for the secondary interest tag; the third weight characterizes the interest degree of each user in the secondary interest tag; determining a fourth weight for said each user based on said first weight, said second weight and said third weight; the fourth weight represents the degree of the each user's preference for the video; matching preset label information of different types of users according to the fourth weight to obtain target label information of a target type; the video recommendation is performed based on the target tag information, the first-level weight and the second-level weight of the user interest tag are considered, the feedback update is timely performed, and the video recommendation effect for the user is better and more accurate.
In this embodiment, a video recommendation apparatus is provided, and fig. 2 is a schematic structural diagram of a video recommendation apparatus according to an embodiment of the present invention, as shown in fig. 2, the apparatus 200 includes: an obtaining unit 201, a first determining unit 202, a second determining unit 203, a matching unit 204 and a recommending unit 205, wherein:
the obtaining unit 201 is configured to obtain a plurality of primary interest tags affecting video interests of a user and a plurality of secondary interest tags corresponding to each of the primary interest tags; the secondary interest label represents the attribute value of the primary interest label;
the first determining unit 202 is configured to determine, according to videos historically browsed by multiple users, a first weight corresponding to a primary interest tag of each of the multiple users, and a second weight and a third weight corresponding to a secondary interest tag under the primary interest tag; the first weight represents the degree of importance of each user on the primary interest tag; the second weight represents the timeliness degree corresponding to the time adopted by each user for the secondary interest tag; the third weight characterizes the interest degree of each user in the secondary interest tag;
the second determining unit 203 is configured to determine a fourth weight of each user based on the first weight, the second weight, and the third weight; the fourth weight represents the degree of the each user's preference for the video;
the matching unit 204 is configured to match preset tag information of different types of users according to the fourth weight, and obtain target tag information of a target type;
the recommending unit 205 is configured to recommend a video based on the target tag information.
In other embodiments, the plurality of primary interest tags includes at least: each video corresponds to a director, actors, genre, drama, distribution area, language, winning award, and source of script.
In other embodiments, the first determining unit 202 is further configured to determine, according to videos historically browsed by multiple users, the number and the maximum word frequency of the primary interest tags corresponding to the videos selected by each user; and determining the first weight according to the number and the maximum word frequency.
In other embodiments, the first determining unit 202 is further configured to obtain, according to videos historically browsed by multiple users, a first time when each user selects that the secondary interest tag corresponding to the video under the primary interest tag is adopted and a second time when the video is initially viewed; determining the second weight according to the first time and the second time.
In other embodiments, the first determining unit 202 is further configured to obtain, according to videos historically viewed by multiple users, a number of times that the secondary interest tag corresponding to the video selected by each user under the primary interest tag is adopted and a total number of videos to be viewed; and determining the third weight according to the times and the total number.
In other embodiments, the second determining unit 203 is further configured to obtain an adjusting factor for adjusting the first weight and a balance factor for measuring the second weight and the third weight distribution; determining a composite weight for each secondary interest tag based on the second weight, the third weight, and the balance factor; determining the fourth weight according to the first weight, the adjustment factor, the second weight, the third weight, and the balance factor.
In other embodiments, the second determining unit 203 is further configured to determine a fifth weight of each secondary interest tag based on the second weight, the third weight and the balance factor; determining the fourth weight according to the first weight, the adjustment factor, and the fifth weight.
The above description of the apparatus embodiments, similar to the above description of the method embodiments, has similar beneficial effects as the method embodiments. For technical details not disclosed in the embodiments of the apparatus according to the invention, reference is made to the description of the embodiments of the method according to the invention for understanding.
It should be noted that, in the embodiment of the present invention, if the video recommendation method is implemented in the form of a software functional module and sold or used as a standalone product, the video recommendation method may also be stored in a computer readable storage medium. With this understanding, technical embodiments of the present invention may be embodied in the form of a software product, which is stored in a storage medium and includes instructions for causing a control server (which may be a personal computer, a server, or a network server) to perform all or part of the methods described in the embodiments of the present invention. And the aforementioned storage medium includes: various media capable of storing program codes, such as a usb disk, a removable hard disk, a Read Only Memory (ROM), a magnetic disk, or an optical disk. Thus, embodiments of the invention are not limited to any specific combination of hardware and software.
Correspondingly, an embodiment of the present invention provides an electronic device, including: a processor and a memory for storing a computer program capable of running on the processor, wherein the processor is configured to execute the steps of the video recommendation method provided by the above embodiments when the computer program is run.
Correspondingly, the embodiment of the invention provides a computer-readable storage medium, on which a computer program is stored, which, when being executed by a processor, realizes the steps in the video recommendation method provided by the above-mentioned embodiment.
Here, it should be noted that: the above description of the storage medium and server embodiments is similar to the description of the method embodiments described above, with similar beneficial effects as the method embodiments. For technical details not disclosed in the embodiments of the storage medium and the server of the present invention, reference is made to the description of the embodiments of the method of the present invention for understanding.
It should be noted that fig. 3 is a schematic diagram of a hardware entity structure of an electronic device in an embodiment of the present invention, and as shown in fig. 3, the hardware entity of the electronic device 300 includes: a processor 301 and a memory 303, optionally, the electronic device 300 may further comprise a communication interface 302.
It will be appreciated that the memory 303 can be either volatile memory or nonvolatile memory, and can include both volatile and nonvolatile memory. Among them, the nonvolatile Memory may be a Read Only Memory (ROM), a Programmable Read Only Memory (PROM), an Erasable Programmable Read-Only Memory (EPROM), an Electrically Erasable Programmable Read-Only Memory (EEPROM), a magnetic random access Memory (FRAM), a Flash Memory (Flash Memory), a magnetic surface Memory, an optical disk, or a Compact Disc Read-Only Memory (CD-ROM); the magnetic surface storage may be disk storage or tape storage. Volatile Memory can be Random Access Memory (RAM), which acts as external cache Memory. By way of illustration and not limitation, many forms of RAM are available, such as Static Random Access Memory (SRAM), Synchronous Static Random Access Memory (SSRAM), Dynamic Random Access Memory (DRAM), Synchronous Dynamic Random Access Memory (SDRAM), Double Data Rate Synchronous Dynamic Random Access Memory (DDRSDRAM), Enhanced Synchronous Dynamic Random Access Memory (ESDRAM), Enhanced Synchronous Dynamic Random Access Memory (Enhanced DRAM), Synchronous Dynamic Random Access Memory (SLDRAM), Direct Memory (DRmb Access), and Random Access Memory (DRAM). The memory 303 described in connection with the embodiments of the invention is intended to comprise, without being limited to, these and any other suitable types of memory.
The method disclosed in the above embodiments of the present invention may be applied to the processor 301, or implemented by the processor 301. The processor 301 may be an integrated circuit chip having signal processing capabilities. In implementation, the steps of the above method may be performed by integrated logic circuits of hardware or instructions in the form of software in the processor 301. The Processor 301 may be a general purpose Processor, a Digital Signal Processor (DSP), or other programmable logic device, discrete gate or transistor logic device, discrete hardware components, or the like. Processor 301 may implement or perform the methods, steps, and logic blocks disclosed in embodiments of the present invention. A general purpose processor may be a microprocessor or any conventional processor or the like. The steps of the method disclosed by the embodiment of the invention can be directly implemented by a hardware decoding processor, or can be implemented by combining hardware and software modules in the decoding processor. The software modules may be located in a storage medium located in the memory 303, and the processor 301 reads the information in the memory 303 and performs the steps of the aforementioned methods in conjunction with its hardware.
In an exemplary embodiment, the electronic Device may be implemented by one or more Application Specific Integrated Circuits (ASICs), DSPs, Programmable Logic Devices (PLDs), Complex Programmable Logic Devices (CPLDs), Field-Programmable Gate arrays (FPGAs), general purpose processors, controllers, Micro Controllers (MCUs), microprocessors (microprocessors), or other electronic components for performing the foregoing methods.
In the embodiments provided in the present invention, it should be understood that the disclosed method and apparatus can be implemented in other ways. The above-described device embodiments are merely illustrative, for example, the division of the unit is only a logical functional division, and there may be other division ways in actual implementation, such as: multiple units or components may be combined, or may be integrated into another observation, or some features may be omitted, or not performed. In addition, the communication connections between the components shown or discussed may be through interfaces, indirect couplings or communication connections of devices or units, and may be electrical, mechanical or other.
The units described as separate parts may or may not be physically separate, and parts displayed as units may or may not be physical units, that is, may be located in one place, or may be distributed on a plurality of network units; some or all of the units can be selected according to actual needs to achieve the purpose of the embodiment.
Those of ordinary skill in the art will understand that: all or part of the steps for realizing the method embodiments can be completed by hardware related to program instructions, the program can be stored in a computer readable storage medium, and the program executes the steps comprising the method embodiments when executed; and the aforementioned storage medium includes: various media that can store program codes, such as a removable Memory device, a Read-Only Memory (ROM), a magnetic disk, or an optical disk.
Alternatively, the integrated unit according to the embodiment of the present invention may be stored in a computer-readable storage medium if it is implemented in the form of a software functional unit and sold or used as a separate product. With this understanding, technical embodiments of the present invention may be embodied in the form of a software product, which is stored in a storage medium and includes instructions for causing an electronic device (which may be a personal computer, a server, or a network device) to perform all or part of the methods described in the embodiments of the present invention. And the aforementioned storage medium includes: a removable storage device, a ROM, a magnetic or optical disk, or other various media that can store program code.
The video recommendation method, the electronic device and the storage medium described in the embodiments of the present invention are only examples of the embodiments of the present invention, but are not limited thereto, and the video recommendation method, the electronic device and the storage medium are all within the scope of the present invention.
It should be appreciated that reference throughout this specification to "one embodiment" or "an embodiment" means that a particular feature, structure or characteristic described in connection with the embodiment is included in at least one embodiment of the present invention. Thus, the appearances of the phrases "in one embodiment" or "in an embodiment" in various places throughout this specification are not necessarily all referring to the same embodiment. Furthermore, the particular features, structures, or characteristics may be combined in any suitable manner in one or more embodiments. It should be understood that, in various embodiments of the present invention, the sequence numbers of the above-mentioned processes do not mean the execution sequence, and the execution sequence of each process should be determined by its function and inherent logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention. The above-mentioned serial numbers of the embodiments of the present invention are merely for description and do not represent the merits of the embodiments.
It should be noted that, in this document, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other like elements in a process, method, article, or apparatus that comprises the element.
The methods disclosed in the several method embodiments provided by the present invention can be combined arbitrarily without conflict to obtain new method embodiments.
Features disclosed in several of the product embodiments provided by the invention may be combined in any combination to yield new product embodiments without conflict.
The features disclosed in the several method or apparatus embodiments provided by the present invention may be combined arbitrarily, without conflict, to arrive at new method embodiments or apparatus embodiments.
The above description is only an embodiment of the present invention, but the scope of the present invention is not limited thereto, and any person skilled in the art can easily conceive of changes or substitutions within the technical scope of the present invention, and all such changes or substitutions are included in the scope of the present invention. Therefore, the protection scope of the present invention shall be subject to the protection scope of the appended claims.

Claims (10)

1. A method for video recommendation, the method comprising:
obtaining a plurality of first-level interest tags influencing the video interest and hobbies of a user and a plurality of second-level interest tags corresponding to each first-level interest tag; the secondary interest label represents the attribute value of the primary interest label;
determining a first weight of a primary interest tag corresponding to each user in a plurality of users and a second weight and a third weight of a secondary interest tag corresponding to the primary interest tag according to videos historically browsed by the users; the first weight represents the degree of importance of each user on the primary interest tag; the second weight represents the timeliness degree corresponding to the time adopted by each user for the secondary interest tag; the third weight characterizes the interest degree of each user in the secondary interest tag;
determining a fourth weight for said each user based on said first weight, said second weight and said third weight; the fourth weight represents the degree of the each user's preference for the video;
matching preset label information of different types of users according to the fourth weight to obtain target label information of a target type;
and recommending videos based on the target label information.
2. The method of claim 1, wherein the plurality of primary interest tags comprises at least: each video corresponds to a director, actors, genre, drama, distribution area, language, winning award, and source of script.
3. The method of claim 1, wherein determining a first weight for each of the plurality of users corresponding to a primary interest tag based on videos historically viewed by the plurality of users comprises:
determining the number and the maximum word frequency of the primary interest labels corresponding to the videos selected by each user according to the videos historically browsed by the users;
and determining the first weight according to the number and the maximum word frequency.
4. The method of claim 1, wherein determining a second weight for each of the plurality of users for a secondary interest tag under the primary interest tag based on videos historically viewed by the plurality of users comprises:
obtaining a first time when each user selects the secondary interest tag corresponding to the video under the primary interest tag and a second time when the video is watched for the first time according to videos historically browsed by a plurality of users;
determining the second weight according to the first time and the second time.
5. The method of claim 1, wherein determining a third weight for each of the plurality of users for a secondary interest tag under the primary interest tag based on videos historically viewed by the plurality of users comprises:
obtaining the times that each user selects the secondary interest tag corresponding to the video under the primary interest tag to be adopted and the total number of watched videos according to videos historically browsed by a plurality of users;
and determining the third weight according to the times and the total number.
6. The method of any of claims 1-5, wherein determining the fourth weight for each user based on the first weight, the second weight, and the third weight comprises:
obtaining an adjusting factor for adjusting the first weight and a balance factor for weighing the second weight and the third weight distribution;
determining a composite weight for each secondary interest tag based on the second weight, the third weight, and the balance factor;
determining the fourth weight according to the first weight, the adjustment factor, the second weight, the third weight, and the balance factor.
7. The method of claim 6, wherein determining the fourth weight from the first weight, the adjustment factor, the second weight, the third weight, and the balance factor comprises:
determining a fifth weight for each secondary interest tag based on the second weight, the third weight, and the balance factor;
determining the fourth weight according to the first weight, the adjustment factor, and the fifth weight.
8. A video recommendation apparatus, characterized in that the apparatus comprises: the device comprises an obtaining unit, a first determining unit, a second determining unit, a matching unit and a recommending unit, wherein:
the obtaining unit is used for obtaining a plurality of first-level interest tags influencing the video interest and hobbies of the user and a plurality of second-level interest tags corresponding to each first-level interest tag; the secondary interest label represents the attribute value of the primary interest label;
the first determining unit is used for determining a first weight corresponding to a primary interest tag of each user in a plurality of users and a second weight and a third weight corresponding to a secondary interest tag under the primary interest tag according to videos historically browsed by the users; the first weight represents the degree of importance of each user on the primary interest tag; the second weight represents the timeliness degree corresponding to the time adopted by each user for the secondary interest tag; the third weight characterizes the interest degree of each user in the secondary interest tag;
the second determining unit is configured to determine a fourth weight of each user based on the first weight, the second weight, and the third weight; the fourth weight represents the degree of the each user's preference for the video;
the matching unit is used for matching preset label information of different types of users according to the fourth weight to obtain target label information of a target type;
and the recommending unit is used for recommending videos based on the target label information.
9. An electronic device, comprising: a processor and a memory for storing a computer program capable of running on the processor,
wherein the processor is adapted to perform the steps of the method of any one of claims 1 to 7 when running the computer program.
10. A computer-readable storage medium, on which a computer program is stored, which, when being executed by a processor, carries out the steps of the method of any one of claims 1 to 7.
CN202111406487.XA 2021-11-24 2021-11-24 Video recommendation method and device, electronic equipment and storage medium Pending CN114168791A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202111406487.XA CN114168791A (en) 2021-11-24 2021-11-24 Video recommendation method and device, electronic equipment and storage medium

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202111406487.XA CN114168791A (en) 2021-11-24 2021-11-24 Video recommendation method and device, electronic equipment and storage medium

Publications (1)

Publication Number Publication Date
CN114168791A true CN114168791A (en) 2022-03-11

Family

ID=80480793

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202111406487.XA Pending CN114168791A (en) 2021-11-24 2021-11-24 Video recommendation method and device, electronic equipment and storage medium

Country Status (1)

Country Link
CN (1) CN114168791A (en)

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2017107453A1 (en) * 2015-12-23 2017-06-29 乐视控股（北京）有限公司 Video content recommendation method, device, and system
CN109543111A (en) * 2018-11-28 2019-03-29 广州虎牙信息科技有限公司 Recommendation information screening technique, device, storage medium and server
WO2019072091A1 (en) * 2017-10-12 2019-04-18 北京京东尚科信息技术有限公司 Method and apparatus for use in determining tags of interest to user
CN111709812A (en) * 2020-06-19 2020-09-25 山东财经大学 E-commerce platform commodity recommendation method and system based on user dynamic classification
CN112989174A (en) * 2019-12-12 2021-06-18 腾讯科技（北京）有限公司 Information recommendation method and device, medium and equipment

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2017107453A1 (en) * 2015-12-23 2017-06-29 乐视控股（北京）有限公司 Video content recommendation method, device, and system
WO2019072091A1 (en) * 2017-10-12 2019-04-18 北京京东尚科信息技术有限公司 Method and apparatus for use in determining tags of interest to user
CN109543111A (en) * 2018-11-28 2019-03-29 广州虎牙信息科技有限公司 Recommendation information screening technique, device, storage medium and server
CN112989174A (en) * 2019-12-12 2021-06-18 腾讯科技（北京）有限公司 Information recommendation method and device, medium and equipment
CN111709812A (en) * 2020-06-19 2020-09-25 山东财经大学 E-commerce platform commodity recommendation method and system based on user dynamic classification

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
夏秀峰;张姝;李晓明;: "一种个性化标签云中的标签排序算法", 沈阳航空航天大学学报, no. 01, 15 February 2011 (2011-02-15) *
许良武;: "基于向量化标签的视频推荐算法研究与实现", 无线互联科技, no. 12, 25 June 2020 (2020-06-25) *

Similar Documents

Publication Publication Date Title
US20220122099A1 (en) Analytical precursor mining for personalized recommendation
US7836001B2 (en) Recommender system with AD-HOC, dynamic model composition
US20180046717A1 (en) Related entities
US8352455B2 (en) Processing a content item with regard to an event and a location
US9171262B2 (en) Directed expertise level-based discovery system, method, and device
US20190171725A1 (en) Building user profiles by relevance feedback
US9727880B2 (en) Predicting user responses
US20140358832A1 (en) Personalized Activity Stream Discovery System, Method, and Device
US20180246899A1 (en) Generate an index for enhanced search based on user interests
US20090006368A1 (en) Automatic Video Recommendation
US20180285774A1 (en) Collaborative personalization via simultaneous embedding of users and their preferences
CN108256093A (en) A kind of Collaborative Filtering Recommendation Algorithm based on the more interest of user and interests change
Nikolakopoulos et al. Trust your neighbors: A comprehensive survey of neighborhood-based methods for recommender systems
US9615136B1 (en) Video classification
US20100250526A1 (en) Search System that Uses Semantic Constructs Defined by Your Social Network
US20190266283A1 (en) Content channel curation
TW201447797A (en) Method and system for multi-phase ranking for content personalization
JP2005135071A (en) Method and device for calculating trust values on purchase
US20120109945A1 (en) Method and system of improving navigation within a set of electronic documents
WO2015025248A2 (en) A system apparatus circuit method and associated computer executable code for hybrid content recommendation
JP2017146926A (en) Information processing device, information processing method and information processing program
JP5416063B2 (en) Content recommendation device, content recommendation program and recording medium thereof
CN114168791A (en) Video recommendation method and device, electronic equipment and storage medium
Chartron et al. General introduction to recommender systems
Hölbling et al. Content-based tag generation to enable a tag-based collaborative tv-recommendation system.

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination