CN110689035B - User classification method, storage medium, electronic device and system - Google Patents
User classification method, storage medium, electronic device and system Download PDFInfo
- Publication number
- CN110689035B CN110689035B CN201810739572.XA CN201810739572A CN110689035B CN 110689035 B CN110689035 B CN 110689035B CN 201810739572 A CN201810739572 A CN 201810739572A CN 110689035 B CN110689035 B CN 110689035B
- Authority
- CN
- China
- Prior art keywords
- classification
- similarity
- modularity
- user
- classifications
- 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.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 31
- 238000004422 calculation algorithm Methods 0.000 claims description 42
- 238000004364 calculation method Methods 0.000 claims description 29
- 238000004590 computer program Methods 0.000 claims description 22
- 230000006399 behavior Effects 0.000 claims description 19
- 238000011156 evaluation Methods 0.000 claims description 9
- 238000010586 diagram Methods 0.000 description 10
- 230000006870 function Effects 0.000 description 4
- 238000012986 modification Methods 0.000 description 4
- 230000004048 modification Effects 0.000 description 4
- 238000005516 engineering process Methods 0.000 description 3
- 239000010749 BS 2869 Class C1 Substances 0.000 description 2
- 239000010750 BS 2869 Class C2 Substances 0.000 description 2
- 230000008901 benefit Effects 0.000 description 2
- 239000002131 composite material Substances 0.000 description 2
- 230000007547 defect Effects 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- 230000009471 action Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 230000008569 process Effects 0.000 description 1
- 230000004044 response Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/22—Matching criteria, e.g. proximity measures
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/02—Marketing; Price estimation or determination; Fundraising
- G06Q30/0201—Market modelling; Market analysis; Collecting market data
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/01—Social networking
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Strategic Management (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Finance (AREA)
- Development Economics (AREA)
- Accounting & Taxation (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Artificial Intelligence (AREA)
- General Business, Economics & Management (AREA)
- Economics (AREA)
- General Engineering & Computer Science (AREA)
- Marketing (AREA)
- Evolutionary Computation (AREA)
- Evolutionary Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Bioinformatics & Computational Biology (AREA)
- Entrepreneurship & Innovation (AREA)
- Life Sciences & Earth Sciences (AREA)
- Computing Systems (AREA)
- Tourism & Hospitality (AREA)
- Primary Health Care (AREA)
- Human Resources & Organizations (AREA)
- General Health & Medical Sciences (AREA)
- Health & Medical Sciences (AREA)
- Game Theory and Decision Science (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention discloses a user classification method, a storage medium, electronic equipment and a system, which relate to the field of big data, and the method comprises the following steps of S1: setting a minimum classification rule, classifying the users to the minimum classification to obtain a live broadcast platform containing a plurality of classifications, selecting two classifications on the live broadcast platform, and entering the step S2; s2: setting the classification after the two classifications are combined as a pre-combination classification, and respectively calculating the modularity of the two classifications and the pre-combination classification; s3: if the sum of the modularity of the two classifications is smaller than the modularity of the pre-merging classification, merging the two classifications, selecting another pair of classifications on the live broadcast platform optionally, and returning to the step S2; if the sum of the modularity of the two classifications is larger than the modularity of the pre-merging classification; optionally selecting another pair of classifications on the live platform, and returning to the step S2; and if the sum of the modularity of any two classifications on the live broadcast platform is less than the modularity of the corresponding pre-merging classification, finishing the classification of the user.
Description
Technical Field
The invention relates to the field of big data, in particular to a user classification method, a storage medium, electronic equipment and a system.
Background
For some web sites, there are various users who can communicate and transmit information for one domain. These users may be interested in specific information, and the website may consume a lot of resources and manpower if it is managed for the interests, habits, etc. of the users one by one. Furthermore, the user has a need to communicate with people with the same interests in addition to the information of interest. Thus, the website needs to group users to help users find their grouping.
This grouping is particularly important for live platforms. Users can be interested in some live broadcast or live broadcast contents, and if a user group with similar interests can be found, the connection among the users can be established through the interest group, so that the social attributes and the user experience of the live broadcast website are enhanced. Therefore, there is a need to find these potential interest groups in a reasonable way.
The common interest group discovery algorithm adopts a clustering technology, but the clustering technology has the defects that the clustering result is uncontrollable, and the dividing reason of the classified groups is difficult to specifically explain. In addition, if the division of the group is too small, the kind of the interest group is too complicated, which is not beneficial to website management and control, and if the division of the group is too large, the personalization for the user will be lost too much, so a user classification method is urgently needed to help the live broadcast platform to perform the user group division.
Disclosure of Invention
Aiming at the defects in the prior art, the invention aims to provide a user classification method, a storage medium, electronic equipment and a system, which can help a live broadcast platform to carry out more reasonable and effective user group division.
In order to achieve the above object, the present invention provides a user classification method, which is characterized in that:
s1: setting a minimum classification rule, classifying the users to the minimum classification to obtain a live broadcast platform containing a plurality of classifications, selecting two classifications on the live broadcast platform, and entering the step S2;
s2: setting the combined classification of the two classifications as a pre-combination classification, and respectively calculating the modularity of the two classifications and the pre-combination classification;
s3: if the sum of the modularity of the two classifications is less than the modularity of the pre-merging classification, merging the two classifications, optionally selecting another pair of classifications on the live broadcast platform, and returning to the step S2;
if the sum of the two classification modularity degrees is greater than the modularity degree of the pre-merging classification; optionally selecting another pair of classifications on the live platform, and returning to the step S2;
and if the sum of the modularity of any two classifications on the live broadcast platform is less than the modularity of the corresponding pre-merging classification, finishing the classification of the user.
On the basis of the technical scheme, the calculation and classification modularity specifically comprises the following steps:
setting any classification in the live broadcast platform as a classification to be evaluated;
calculating the similarity between users in a to-be-evaluated classification according to a preset user similarity algorithm, and accumulating to obtain a first similarity;
calculating the similarity between users of the whole live broadcast platform according to a preset user similarity algorithm, and accumulating to obtain a second similarity;
according to a preset user similarity algorithm, calculating the similarity between the users in the to-be-evaluated classification and the users in the to-be-evaluated classification on the live broadcast platform, and accumulating to obtain a third similarity;
and dividing the first similarity by the second similarity to obtain a first percentage, dividing the third similarity by the second similarity to obtain a square of a quotient serving as a second percentage, and subtracting the second percentage from the first percentage to obtain the modularity of the classification to be evaluated.
On the basis of the technical scheme, the specific calculation formula of the classification modularity is as follows:
wherein Q issRepresenting the modularity of the classification; k represents the number of categories in the current live platform; ISiRepresenting a first similarity; TS represents a second similarity; DS (direct sequence) systemiIndicating a third degree of similarity.
On the basis of the technical scheme, the specific formula of the preset similarity algorithm is as follows:
wherein s (u, v) represents the similarity between the user u and the user v, τ (u) represents the set of live broadcasting rooms watched by the user u, τ (v) represents the set of live broadcasting rooms watched by the user v, τ (u) # τ (v) represents the set of live broadcasting rooms watched by the user u and the user v together, w (u, x) represents the characteristic value of the live broadcasting room x watched by the user u in a preset time period, and w (v, x) represents the characteristic value of the live broadcasting room x watched by the user v in the preset time period;
the characteristic value is obtained by acquiring the behavior characteristics of the user in the corresponding live broadcast room and scoring according to preset behavior evaluation indexes.
On the basis of the above technical solution, the minimum classification rule is:
the method comprises the steps of obtaining N item values of a user on a live broadcast platform, setting weighting for each item value, accumulating to obtain a comprehensive value, and classifying the user into a preset classification when the comprehensive value is larger than a preset threshold value, wherein N is an integer larger than 0.
On the basis of the technical scheme, the time length ratio of each live broadcast room watched by the user is a project numerical value, each live broadcast room on the live broadcast platform is of one type, when the time length ratio of any live broadcast room watched by the user exceeds 50% in a preset time period, the user is classified to the live broadcast room, and the time length ratio is the percentage of the time length of one live broadcast room station watched by the user to the total time length of all live broadcast rooms.
The present invention also provides a storage medium having a computer program stored thereon, characterized in that: the computer program when executed by a processor implementing the steps of:
s1: setting a minimum classification rule, classifying the users to the minimum classification to obtain a live broadcast platform containing a plurality of classifications, selecting two classifications on the live broadcast platform, and entering the step S2;
s2: setting the combined classification of the two classifications as a pre-combination classification, and respectively calculating the modularity of the two classifications and the pre-combination classification;
s3: if the sum of the modularity of the two classifications is less than the modularity of the pre-merging classification, merging the two classifications, optionally selecting another pair of classifications on the live broadcast platform, and returning to the step S2;
if the sum of the two classification modularity degrees is greater than the modularity degree of the pre-merging classification; optionally selecting another pair of classifications on the live platform, and returning to the step S2;
and if the sum of the modularity of any two classifications on the live broadcast platform is less than the modularity of the corresponding pre-merging classification, finishing the classification of the user.
On the basis of the technical scheme, the calculation and classification modularity specifically comprises the following steps:
setting any classification in the live broadcast platform as a classification to be evaluated;
calculating the similarity between users in a to-be-evaluated classification according to a preset user similarity algorithm, and accumulating to obtain a first similarity;
calculating the similarity between users of the whole live broadcast platform according to a preset user similarity algorithm, and accumulating to obtain a second similarity;
according to a preset user similarity algorithm, calculating the similarity between the users in the to-be-evaluated classification and the users in the to-be-evaluated classification on the live broadcast platform, and accumulating to obtain a third similarity;
and dividing the first similarity by the second similarity to obtain a first percentage, dividing the third similarity by the second similarity to obtain a square of a quotient serving as a second percentage, and subtracting the second percentage from the first percentage to obtain the modularity of the classification to be evaluated.
On the basis of the technical scheme, the specific calculation formula of the classification modularity is as follows:
wherein Q issRepresenting the modularity of the classification; k represents the number of categories in the current live platform; ISiRepresenting a first similarity; TS represents a second similarity; DS (direct sequence)iIndicating a third degree of similarity.
On the basis of the technical scheme, the specific formula of the preset similarity algorithm is as follows:
s (u, v) represents the similarity between a user u and a user v, τ (u) represents the set of live broadcasting rooms watched by the user u, τ (v) represents the set of live broadcasting rooms watched by the user v, τ (u) # τ (v) represents the set of live broadcasting rooms watched by the user u and the user v together, w (u, x) represents a characteristic value of the live broadcasting room x watched by the user u in a preset time period, and w (v, x) represents a characteristic value of the live broadcasting room x watched by the user v in the preset time period;
the characteristic value is obtained by acquiring the behavior characteristics of the user in the corresponding live broadcast room and scoring according to preset behavior evaluation indexes.
On the basis of the above technical solution, the minimum classification rule is:
the method comprises the steps of obtaining N item values of a user on a live broadcast platform, setting weighting for each item value, accumulating to obtain a comprehensive value, and classifying the user into a preset classification when the comprehensive value is larger than a preset threshold value, wherein N is an integer larger than 0.
On the basis of the technical scheme, the time length ratio of each live broadcast room watched by the user is a project numerical value, each live broadcast room on the live broadcast platform is of one type, when the time length ratio of any live broadcast room watched by the user exceeds 50% in a preset time period, the user is classified to the live broadcast room, and the time length ratio is the percentage of the time length of one live broadcast room station watched by the user to the total time length of all live broadcast rooms.
The present invention also provides an electronic device comprising a memory and a processor, the memory having stored thereon a computer program that runs on the processor: the processor, when executing the computer program, implements the steps of:
s1: setting a minimum classification rule, classifying the users to the minimum classification to obtain a live broadcast platform containing a plurality of classifications, selecting two classifications on the live broadcast platform, and entering the step S2;
s2: setting the combined classification of the two classifications as a pre-combination classification, and respectively calculating the modularity of the two classifications and the pre-combination classification;
s3: if the sum of the modularity of the two classifications is less than the modularity of the pre-merging classification, merging the two classifications, optionally selecting another pair of classifications on the live broadcast platform, and returning to the step S2;
if the sum of the two classification modularity degrees is greater than the modularity degree of the pre-merging classification; optionally selecting another pair of classifications on the live platform, and returning to the step S2;
and if the sum of the modularity of any two classifications on the live broadcast platform is less than the modularity of the corresponding pre-merging classification, finishing the classification of the user.
On the basis of the technical scheme, the calculation and classification modularity specifically comprises the following steps:
setting any classification in the live broadcast platform as a classification to be evaluated;
calculating the similarity between users in a to-be-evaluated classification according to a preset user similarity algorithm, and accumulating to obtain a first similarity;
calculating the similarity between users of the whole live broadcast platform according to a preset user similarity algorithm, and accumulating to obtain a second similarity;
according to a preset user similarity algorithm, calculating the similarity between the users in the to-be-evaluated classification and the users in the to-be-evaluated classification on the live broadcast platform, and accumulating to obtain a third similarity;
and dividing the first similarity by the second similarity to obtain a first percentage, dividing the third similarity by the second similarity to obtain a square of a quotient serving as a second percentage, and subtracting the second percentage from the first percentage to obtain the modularity of the classification to be evaluated.
On the basis of the technical scheme, the specific calculation formula of the classification modularity is as follows:
wherein Q issRepresenting the modularity of the classification; k represents the number of categories in the current live platform; ISiRepresenting a first similarity; TS represents a second similarity; DS (direct sequence)iIndicating a third degree of similarity.
On the basis of the technical scheme, the specific formula of the preset similarity algorithm is as follows:
s (u, v) represents the similarity between a user u and a user v, τ (u) represents the set of live broadcasting rooms watched by the user u, τ (v) represents the set of live broadcasting rooms watched by the user v, τ (u) # τ (v) represents the set of live broadcasting rooms watched by the user u and the user v together, w (u, x) represents a characteristic value of the live broadcasting room x watched by the user u in a preset time period, and w (v, x) represents a characteristic value of the live broadcasting room x watched by the user v in the preset time period;
the characteristic value is obtained by acquiring the behavior characteristics of the user in the corresponding live broadcast room and scoring according to preset behavior evaluation indexes.
On the basis of the above technical solution, the minimum classification rule is:
the method comprises the steps of obtaining N item values of a user on a live broadcast platform, setting weighting for each item value, accumulating to obtain a comprehensive value, and classifying the user into a preset classification when the comprehensive value is larger than a preset threshold value, wherein N is an integer larger than 0.
On the basis of the technical scheme, the time length ratio of each live broadcast room watched by the user is a project numerical value, each live broadcast room on the live broadcast platform is of one type, when the time length ratio of any live broadcast room watched by the user exceeds 50% in a preset time period, the user is classified to the live broadcast room, and the time length ratio is the percentage of the time length of one live broadcast room station watched by the user to the total time length of all live broadcast rooms.
The invention also provides a user classification system, which comprises an initial classification unit, a modularity calculation unit and a merging judgment unit:
the system comprises an initial classification unit, a modularity calculating unit and a control unit, wherein the initial classification unit is used for classifying users to a minimum classification according to a minimum classification rule to obtain a live broadcast platform containing a plurality of classifications, and two classifications on an optional live broadcast platform are sent to the modularity calculating unit;
the modularity calculating unit is used for setting the two classifications after the classification combination as a pre-combination classification and respectively calculating the modularity of the two classifications and the pre-combination classification;
and the combination judging unit is used for combining the two classifications when the sum of the modularity of the two classifications is less than the modularity of the pre-combination classification to obtain the live broadcasting platform after the two classifications are combined, sending a pair of classifications on the optional live broadcasting platform to the modularity calculating unit, sending the other pair of classifications on the optional live broadcasting platform to the modularity calculating unit when the sum of the modularity of the two classifications is greater than the modularity of the pre-combination classification, and finishing the classification of the user when the sum of the modularity of any two classifications on the live broadcasting platform is less than the modularity of the corresponding pre-combination classification.
On the basis of the technical scheme, the preset similarity algorithm has the specific formula as follows:
s (u, v) represents the similarity between a user u and a user v, τ (u) represents the set of live broadcasting rooms watched by the user u, τ (v) represents the set of live broadcasting rooms watched by the user v, τ (u) # τ (v) represents the set of live broadcasting rooms watched by the user u and the user v together, w (u, x) represents a characteristic value of the live broadcasting room x watched by the user u in a preset time period, and w (v, x) represents a characteristic value of the live broadcasting room x watched by the user v in the preset time period;
the characteristic value is obtained by acquiring the behavior characteristics of the user in the corresponding live broadcast room and scoring according to preset behavior evaluation indexes.
On the basis of the technical scheme, the calculation and classification modularity specifically comprises the following steps:
setting any classification in the live broadcast platform as a classification to be evaluated;
calculating the similarity between users in a to-be-evaluated classification according to a preset user similarity algorithm, and accumulating to obtain a first similarity;
calculating the similarity between users of the whole live broadcast platform according to a preset user similarity algorithm, and accumulating to obtain a second similarity;
according to a preset user similarity algorithm, calculating the similarity between the users in the to-be-evaluated classification and the users in the to-be-evaluated classification on the live broadcast platform, and accumulating to obtain a third similarity;
and dividing the first similarity by the second similarity to obtain a first percentage, dividing the third similarity by the second similarity to obtain a square of a quotient serving as a second percentage, and subtracting the second percentage from the first percentage to obtain the modularity of the classification to be evaluated.
On the basis of the technical scheme, the specific calculation formula of the classification modularity is as follows:
wherein Q issRepresenting the modularity of the classification; k represents the number of categories in the current live platform; ISiRepresenting a first similarity; TS represents a second similarity; DS (direct sequence)iIndicating a third degree of similarity.
On the basis of the above technical solution, the minimum classification rule is:
the method comprises the steps of obtaining N item values of a user on a live broadcast platform, setting weighting for each item value, accumulating to obtain a comprehensive value, and classifying the user into a preset classification when the comprehensive value is larger than a preset threshold value, wherein N is an integer larger than 0.
On the basis of the technical scheme, the time length ratio of each live broadcast room watched by the user is a project numerical value, each live broadcast room on the live broadcast platform is of one type, when the time length ratio of any live broadcast room watched by the user exceeds 50% in a preset time period, the user is classified to the live broadcast room, and the time length ratio is the percentage of the time length of one live broadcast room station watched by the user to the total time length of all live broadcast rooms.
Compared with the prior art, the invention has the advantages that:
the invention presets an algorithm to calculate the modularity of the live broadcast platform, can measure the connection strength between users of the same classification after classification on the live broadcast platform, then carries out minimum classification on the users on the live broadcast platform, and uses a recursive circulation mode to continuously and circularly combine the classifications which can be combined on the live broadcast platform until the classification combination reduces the modularity of the live broadcast platform, namely reduces the connection between the users, and at the moment, when the live broadcast platform is the most reasonable, the classification can be accurately and automatically carried out.
Drawings
In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings corresponding to the embodiments are briefly introduced below, and it is obvious that the drawings in the following description are some embodiments of the present invention, and it is obvious for those skilled in the art that other drawings can be obtained according to these drawings without creative efforts.
Fig. 1 is a schematic structural diagram/an exploded view/a front view/a left view/a right view/a top view/a bottom view/of a user classification method flowchart according to an embodiment of the present invention;
fig. 2 is a schematic structural diagram of a user classification system according to an embodiment of the present invention.
In the figure: 1-an initial classification unit, 2-a modularity degree calculation unit and 3-a merging judgment unit.
Detailed Description
Embodiments of the present invention will be described in further detail below with reference to the accompanying drawings.
Referring to fig. 1, embodiments of the present invention provide a user classification method, a storage medium, an electronic device, and a system, which can more reasonably and effectively classify and classify users by calculating, merging, and judging the modularity of live broadcast platform classification
As shown in fig. 1, in order to achieve the above technical effects, the general idea of the present application is as follows:
s1: and setting a minimum classification rule, classifying the users to the minimum classification to obtain a live broadcast platform containing a plurality of classifications, selecting two classifications on the live broadcast platform, and entering the step S2.
There are three ways to classify users specifically, one is to take the user as a whole, and split continuously until reasonable classification, and this classification way always carries a split basis or label for each split, and the last classification is difficult to specify the basis/label for system automatic search. One is to classify users to a medium degree first, then split the oversize classification, merge the smaller classifications, but the user classification is performed according to specific meanings, not the number, and the system is difficult to judge how large the classification corresponding to a particular meaning is, which needs to be split, how small it is, which needs to be merged. The other is to find the minimum label, assemble the user into the minimum class, and then merge continuously according to whether the merging is reasonable or not. Step S1 is to perform minimum classification first, distribute users to the minimum classification and converge users to the minimum classification first on the live broadcast platform, and prepare for subsequent merging
S2: setting the combined classification of the two classifications as a pre-combination classification, and respectively calculating the modularity of the two classifications and the pre-combination classification;
it should be noted that, an algorithm of modularity is preset for the classification on the live broadcast platform, and the algorithm calculates the similarity of users in each classification on the live broadcast platform, where the modularity is high when the similarity of users is high in the live broadcast platform, and the modularity is low when the similarity of users is low.
Through the step S1, the categories existing on the live broadcast platform are all the minimum categories, and then the minimum categories need to be combined continuously, and a preset algorithm is used to judge whether the categories are reasonable, so that the modularity before and after combination is calculated at first in the step S2.
S3: if the sum of the modularity of the two classifications is less than the modularity of the pre-merging classification, merging the two classifications, optionally selecting another pair of classifications on the live broadcast platform, and returning to the step S2;
if the sum of the two classification modularity degrees is greater than the modularity degree of the pre-merging classification; optionally selecting another pair of classifications on the live platform, and returning to the step S2;
and if the sum of the modularity of any two classifications on the live broadcast platform is less than the modularity of the corresponding pre-merging classification, finishing the classification of the user.
After the modularity of the two modules is calculated, step S3 further performs the next action according to the comparison of the modularity, and if the sum of the modularity of the two classifications is less than the modularity of the pre-merging classification, the two classifications are merged to form a live broadcast platform with a new classification, otherwise, the original higher modularity is not merged and maintained, and step S2 is returned to calculate and judge whether there is a possibility of merging other classifications until the classification on the live broadcast platform reaches the highest modularity, that is, the similarity between users of each classification on the live broadcast platform is relatively high.
In order to better understand the technical scheme, the following detailed description is made in conjunction with the specific embodiments.
Example one
The embodiment of the invention provides a method for classifying users, which comprises the following specific steps:
s1: setting a minimum classification rule to classify the users to the minimum classification to obtain a live broadcast platform containing a plurality of classifications, selecting two classifications on the live broadcast platform, and entering the step S2
As a preferred embodiment, the minimum classification rule used in the method is to obtain N item values of the user on the live broadcast platform, set a weight for each item value, add up to obtain a composite value, and classify the user into a preset classification when the composite value is greater than a preset threshold, where N is an integer greater than 0.
It should be noted that, when there are multiple aspects of the user classification, such as interested tags, frequently watched live broadcast rooms, types of bullet screens, and the like, for the live broadcast platform, different aspects correspond to different statistics and classification requirements, and therefore, the items include various parameters and preferences of the user on the live broadcast platform, as long as the user can be classified into various categories in a wider range.
For example, the time length of watching the live broadcast rooms by the user is classified, and if the time length of watching any live broadcast room exceeds 50% in a preset time period, the user is classified to the live broadcast room classification, wherein the time length is the percentage of the time length of watching one live broadcast room station by the user to the total time length of watching all live broadcast rooms.
S2: setting the combined classification of the two classifications as a pre-combination classification, and respectively calculating the modularity of the two classifications and the pre-combination classification;
as a preferred embodiment, the step of calculating the modularity of the classification in step S2 specifically includes the steps of:
setting any classification in the live broadcast platform as a classification to be evaluated;
calculating the similarity between users in a to-be-evaluated classification according to a preset user similarity algorithm, and accumulating to obtain a first similarity;
calculating the similarity between users of the whole live broadcast platform according to a preset user similarity algorithm, and accumulating to obtain a second similarity;
according to a preset user similarity algorithm, calculating the similarity between the users in the to-be-evaluated classification and the users in the to-be-evaluated classification on the live broadcast platform, and accumulating to obtain a third similarity;
and dividing the first similarity by the second similarity to obtain a first percentage, dividing the third similarity by the second similarity to obtain a square of a quotient serving as a second percentage, and subtracting the second percentage from the first percentage to obtain the modularity of the classification to be evaluated.
Specifically, the specific calculation formula of the classification modularity is as follows:
wherein Q issRepresenting the modularity of the classification; k represents the number of categories in the current live platform; ISiRepresenting a first similarity; TS represents a second similarity; DS (direct sequence)iIndicating a third degree of similarity.
Further, the preset similarity algorithm has a specific formula as follows:
s (u, v) represents the similarity between a user u and a user v, τ (u) represents the set of live broadcasting rooms watched by the user u, τ (v) represents the set of live broadcasting rooms watched by the user v, τ (u) # τ (v) represents the set of live broadcasting rooms watched by the user u and the user v together, w (u, x) represents a characteristic value of the live broadcasting room x watched by the user u in a preset time period, and w (v, x) represents a characteristic value of the live broadcasting room x watched by the user v in the preset time period;
the characteristic value is obtained by acquiring the behavior characteristics of the user in the corresponding live broadcast room and scoring according to preset behavior evaluation indexes.
It should be noted that the calculation of the modularity requires similarity, and therefore an algorithm is required to reflect the similarity between users through numerical values, and to reflect the individuality of the user on the live broadcast platform in terms of numerical values by aiming at different users to watch different characteristic values of the live broadcast room within a preset time period, that is, behavior characteristics of the user in the corresponding live broadcast room, and scoring according to preset behavior evaluation indexes to obtain numerical values, and the algorithm performs specific comprehensive calculation by using the individualized numerical values of the two users as parameters. When the behaviors of the users in the live broadcast room are consistent or similar, the numerical value is higher, otherwise, the numerical value is lower, and the similarity of the users can be definitely reflected.
For example, class C1With users a and b, class C2There are users c and d, and the similarity between them is:
sac＝0.5
sad＝0.1
sbc＝0.2
sbd＝0.3
in addition, in class C1And C2In addition to the users e, and C1And C2The similarity between users of (a) is:
sea＝0.3,seb＝0.4,sec＝0.6,sed＝0.3
therefore, the first similarity, the second similarity and the third similarity corresponding to the parameters of the above-mentioned classified modularity calculation formula are respectively:
US12＝0.5+0.1+0.2+0.3＝1.1
DS1＝0.5+0.1+0.2+0.3+0.3+0.4＝1.8
DS2＝0.5+0.1+0.2+0.3+0.6+0.3＝2
TS＝0.5+0.1+0.2+0.3+0.3+0.4+0.6+0.3＝2.7
will classify C1And C2Combining may result in a modularity gain Δ QsComprises the following steps:
ΔQs< 0 therefore Category C1And C2Cannot be combined.
Example two
Referring to fig. 2, an embodiment of the present invention provides a user classification system, which includes:
the system comprises an initial classification unit 1, a modularity calculating unit and a control unit, wherein the initial classification unit is used for classifying users to a minimum classification according to a minimum classification rule to obtain a live broadcast platform containing a plurality of classifications, and two classifications on an optional live broadcast platform are sent to the modularity calculating unit;
the modularity calculating unit 2 is used for setting the two classifications after the classification combination as a pre-combination classification, and respectively calculating the modularity of the two classifications and the pre-combination classification;
and the merging judgment unit 3 is used for merging the two classifications when the sum of the two classification modularity degrees is less than the modularity degree of the pre-merging classification, so as to obtain the live broadcast platform after the two classifications are merged, sending a pair of classifications on the optional live broadcast platform to the modularity degree calculation unit, sending the other pair of classifications on the optional live broadcast platform to the modularity degree calculation unit when the sum of the two classification modularity degrees is greater than the modularity degree of the pre-merging classification, and finishing the classification of the user when the sum of any two classification modularity degrees on the live broadcast platform is less than the modularity degree of the corresponding pre-merging classification.
As a preferred embodiment, the specific formula of the preset similarity algorithm is as follows:
s (u, v) represents the similarity between a user u and a user v, τ (u) represents the set of live broadcasting rooms watched by the user u, τ (v) represents the set of live broadcasting rooms watched by the user v, τ (u) # τ (v) represents the set of live broadcasting rooms watched by the user u and the user v together, w (u, x) represents a characteristic value of the live broadcasting room x watched by the user u in a preset time period, and w (v, x) represents a characteristic value of the live broadcasting room x watched by the user v in the preset time period;
the characteristic value is obtained by acquiring the behavior characteristics of the user in the corresponding live broadcast room and scoring according to preset behavior evaluation indexes.
Various modifications and specific examples in the foregoing method embodiments are also applicable to the system of the present embodiment, and the detailed description of the method is clear to those skilled in the art, so that the detailed description is omitted here for the sake of brevity.
Based on the same inventive concept, the present application provides a third embodiment, which has the following specific implementation modes:
EXAMPLE III
In response to the user classifying method, the present invention further provides a storage medium, where a computer program is stored on the storage medium, and the computer program implements the following steps when executed by a processor.
S1: setting a minimum classification rule, classifying the users to the minimum classification to obtain a live broadcast platform containing a plurality of classifications, selecting two classifications on the live broadcast platform, and entering the step S2;
s2: setting the combined classification of the two classifications as a pre-combination classification, and respectively calculating the modularity of the two classifications and the pre-combination classification;
s3: if the sum of the modularity of the two classifications is less than the modularity of the pre-merging classification, merging the two classifications, optionally selecting another pair of classifications on the live broadcast platform, and returning to the step S2;
if the sum of the two classification modularity degrees is greater than the modularity degree of the pre-merging classification; optionally selecting another pair of classifications on the live platform, and returning to the step S2;
and if the sum of the modularity of any two classifications on the live broadcast platform is less than the modularity of the corresponding pre-merging classification, finishing the classification of the user.
The storage medium includes various media capable of storing program codes, such as a usb disk, a removable hard disk, a ROM (Read-Only Memory), a RAM (Random Access Memory), a magnetic disk, or an optical disk. These stored computer programs, when executed, implement the methods of the above-described embodiments.
Based on the same inventive concept, the present application provides a fourth embodiment, which has the following specific implementation:
example four
Corresponding to the user classification method, the invention further provides an electronic device, in which a computer program is stored, and when the computer program is executed by a processor, the following steps are implemented:
s1: setting a minimum classification rule, classifying the users to the minimum classification to obtain a live broadcast platform containing a plurality of classifications, selecting two classifications on the live broadcast platform, and entering the step S2;
s2: setting the combined classification of the two classifications as a pre-combination classification, and respectively calculating the modularity of the two classifications and the pre-combination classification;
s3: if the sum of the modularity of the two classifications is less than the modularity of the pre-merging classification, merging the two classifications, optionally selecting another pair of classifications on the live broadcast platform, and returning to the step S2;
if the sum of the two classification modularity degrees is greater than the modularity degree of the pre-merging classification; optionally selecting another pair of classifications on the live platform, and returning to the step S2;
and if the sum of the modularity of any two classifications on the live broadcast platform is less than the modularity of the corresponding pre-merging classification, finishing the classification of the user.
It should be noted that the electronic device includes a memory and a processor, the memory stores a computer program running on the processor, and the processor implements the method of the above-mentioned embodiment when executing the computer program.
Generally speaking, the user classification method, the storage medium, the electronic device and the system provided by the embodiment of the invention can combine the user classification results by comparing and calculating the modularity change before and after the user classification, and can combine the user on the live broadcast platform from the minimum classification to the classification when the modularity of the live broadcast platform reaches the maximum degree compared with the traditional technology and the like, namely the classification combination does not benefit the classification any more, but can complete the combination of the live broadcast platform before combining two user groups which do not want to be related, thereby being capable of helping the live broadcast platform to carry out more reasonable and effective user group division.
As will be appreciated by one skilled in the art, embodiments of the present invention may be provided as a method, system, or computer program product. Accordingly, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present invention may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, optical storage, and the like) having computer-usable program code embodied therein.
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.
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 (8)
1. A method of classifying a user, comprising:
s1: setting a minimum classification rule, classifying the users to the minimum classification to obtain a live broadcast platform containing a plurality of classifications, selecting two classifications on the live broadcast platform, and entering the step S2;
s2: setting the combined classification of the two classifications as a pre-combination classification, and respectively calculating the modularity of the two classifications and the pre-combination classification;
s3: if the sum of the modularity of the two classifications is smaller than the modularity of the pre-merging classification, merging the two classifications, selecting another pair of classifications on the live broadcast platform optionally, and returning to the step S2;
if the sum of the two classification modularity degrees is greater than the modularity degree of the pre-merging classification; optionally selecting another pair of classifications on the live platform, and returning to the step S2;
if the sum of the modularity of any two classifications on the live broadcast platform is less than the modularity of the corresponding pre-merging classification, finishing the classification of the users;
the calculation of the modularity of the classification specifically comprises the following steps:
setting any classification in the live broadcast platform as a classification to be evaluated;
calculating the similarity between users in a to-be-evaluated classification according to a preset user similarity algorithm, and accumulating to obtain a first similarity;
calculating the similarity between users of the whole live broadcast platform according to a preset user similarity algorithm, and accumulating to obtain a second similarity;
according to a preset user similarity algorithm, calculating the similarity between the users in the to-be-evaluated classification and the users in the to-be-evaluated classification on the live broadcast platform, and accumulating to obtain a third similarity;
dividing the first similarity by the second similarity to obtain a first percentage, dividing the third similarity by the second similarity to obtain a quotient, taking the square of the quotient as a second percentage, and subtracting the second percentage from the first percentage to obtain the modularity of the classification to be evaluated;
the specific calculation formula of the classification modularity is as follows:
wherein Q issRepresenting the modularity of the classification; k represents the number of categories in the current live platform; ISiRepresenting a first similarity; TS represents a second similarity; DS (direct sequence)iIndicating a third degree of similarity.
2. The method for classifying users according to claim 1, wherein said predetermined similarity algorithm has a specific formula:
s (u, v) represents the similarity between a user u and a user v, τ (u) represents the set of live broadcasting rooms watched by the user u, τ (v) represents the set of live broadcasting rooms watched by the user v, τ (u) # τ (v) represents the set of live broadcasting rooms watched by the user u and the user v together, w (u, x) represents a characteristic value of the live broadcasting room x watched by the user u in a preset time period, and w (v, x) represents a characteristic value of the live broadcasting room x watched by the user v in the preset time period;
the characteristic value is obtained by acquiring the behavior characteristics of the user in the corresponding live broadcast room and scoring according to preset behavior evaluation indexes.
3. The method of claim 1, wherein the minimum classification rule is:
the method comprises the steps of obtaining N item values of a user on a live broadcast platform, setting weighting for each item value, accumulating to obtain a comprehensive value, and classifying the user into a preset classification when the comprehensive value is larger than a preset threshold value, wherein N is an integer larger than 0.
4. A method for classifying a user according to claim 3, characterized in that: the time length ratio of each live broadcast room watched by a user is a project value, each live broadcast room on the live broadcast platform is of one type, when the time length ratio of any live broadcast room watched by the user is more than 50%, the user is classified to the live broadcast room in a preset time period, and the time length ratio is the percentage of the time length of one live broadcast room station watched by the user to the total time length of all live broadcast rooms watched by the user.
5. A storage medium having a computer program stored thereon, characterized in that: the computer program when executed by a processor implementing the steps of:
s1: setting a minimum classification rule, classifying the users to the minimum classification to obtain a live broadcast platform containing a plurality of classifications, selecting two classifications on the live broadcast platform, and entering the step S2;
s2: setting the combined classification of the two classifications as a pre-combination classification, and respectively calculating the modularity of the two classifications and the pre-combination classification;
s3: if the sum of the modularity of the two classifications is less than the modularity of the pre-merging classification, merging the two classifications, optionally selecting another pair of classifications on the live broadcast platform, and returning to the step S2;
if the sum of the two classification modularity degrees is greater than the modularity degree of the pre-merging classification; optionally selecting another pair of classifications on the live platform, and returning to the step S2;
if the sum of the modularity of any two classifications on the live broadcast platform is less than the modularity of the corresponding pre-merging classification, finishing the classification of the users;
the calculation of the modularity of the classification specifically comprises the following steps:
setting any classification in the live broadcast platform as a classification to be evaluated;
calculating the similarity between users in a to-be-evaluated classification according to a preset user similarity algorithm, and accumulating to obtain a first similarity;
calculating the similarity between users of the whole live broadcast platform according to a preset user similarity algorithm, and accumulating to obtain a second similarity;
according to a preset user similarity algorithm, calculating the similarity between the users in the to-be-evaluated classification and the users in the to-be-evaluated classification on the live broadcast platform, and accumulating to obtain a third similarity;
dividing the first similarity by the second similarity to obtain a first percentage, dividing the third similarity by the second similarity to obtain a square of a quotient serving as a second percentage, and subtracting the second percentage from the first percentage to obtain the modularity of the to-be-evaluated classification;
the specific calculation formula of the classified modularity is as follows:
wherein Q issRepresenting the modularity of the classification; k represents the number of categories in the current live platform; ISiRepresenting a first similarity; TS represents a second similarity; DS (direct sequence)iIndicating a third degree of similarity.
6. An electronic device comprising a memory and a processor, the memory having stored thereon a computer program for execution on the processor, the electronic device comprising: the processor, when executing the computer program, implements the steps of:
s1: setting a minimum classification rule, classifying the users to the minimum classification to obtain a live broadcast platform containing a plurality of classifications, selecting two classifications on the live broadcast platform, and entering the step S2;
s2: setting the combined classification of the two classifications as a pre-combination classification, and respectively calculating the modularity of the two classifications and the pre-combination classification;
s3: if the sum of the modularity of the two classifications is less than the modularity of the pre-merging classification, merging the two classifications, optionally selecting another pair of classifications on the live broadcast platform, and returning to the step S2;
if the sum of the two classification modularity degrees is greater than the modularity degree of the pre-merging classification; optionally selecting another pair of classifications on the live platform, and returning to the step S2;
if the sum of the modularity of any two classifications on the live broadcast platform is less than the modularity of the corresponding pre-merging classification, finishing the classification of the users;
the calculation of the modularity of the classification specifically comprises the following steps:
setting any classification in the live broadcast platform as a classification to be evaluated;
calculating the similarity between users in a to-be-evaluated classification according to a preset user similarity algorithm, and accumulating to obtain a first similarity;
calculating the similarity between users of the whole live broadcast platform according to a preset user similarity algorithm, and accumulating to obtain a second similarity;
according to a preset user similarity algorithm, calculating the similarity between the users in the to-be-evaluated classification and the users in the to-be-evaluated classification on the live broadcast platform, and accumulating to obtain a third similarity;
dividing the first similarity by the second similarity to obtain a first percentage, dividing the third similarity by the second similarity to obtain a square of a quotient serving as a second percentage, and subtracting the second percentage from the first percentage to obtain the modularity of the to-be-evaluated classification;
the specific calculation formula of the classification modularity is as follows:
wherein Q issRepresenting the modularity of the classification; k represents the number of categories in the current live platform; ISiRepresenting a first similarity; TS represents a second similarity; DS (direct sequence)iIndicating a third degree of similarity.
7. A user classification system is characterized by comprising an initial classification unit, a modularity calculation unit and a merging judgment unit:
the system comprises an initial classification unit (1) and a modularity degree calculation unit, wherein the initial classification unit is used for classifying users to a minimum classification according to a minimum classification rule to obtain a live broadcast platform containing a plurality of classifications, and two classifications on any live broadcast platform are sent to the modularity degree calculation unit;
the modularity calculating unit (2) is used for setting the two classifications after the classification combination as a pre-combination classification, and respectively calculating the modularity of the two classifications and the pre-combination classification;
a merging judgment unit (3) for merging the two classifications when the sum of the two classification modularity degrees is less than the modularity degree of the pre-merging classification, so as to obtain a live broadcast platform after the two classifications are merged, sending a pair of classifications on an optional live broadcast platform to a modularity degree calculation unit, sending the other pair of classifications on the optional live broadcast platform to the modularity degree calculation unit when the sum of the two classification modularity degrees is greater than the modularity degree of the pre-merging classification, and finishing the classification of the user when the sum of any two classification modularity degrees on the live broadcast platform is less than the modularity degree of the corresponding pre-merging classification;
the calculation of the modularity of the classification specifically comprises the following steps:
setting any classification in the live broadcast platform as a classification to be evaluated;
calculating the similarity between users in a to-be-evaluated classification according to a preset user similarity algorithm, and accumulating to obtain a first similarity;
calculating the similarity between users of the whole live broadcast platform according to a preset user similarity algorithm, and accumulating to obtain a second similarity;
according to a preset user similarity algorithm, calculating the similarity between the users in the to-be-evaluated classification and the users in the to-be-evaluated classification on the live broadcast platform, and accumulating to obtain a third similarity;
dividing the first similarity by the second similarity to obtain a first percentage, dividing the third similarity by the second similarity to obtain a square of a quotient serving as a second percentage, and subtracting the second percentage from the first percentage to obtain the modularity of the to-be-evaluated classification;
the specific calculation formula of the classification modularity is as follows:
wherein Q issRepresenting the modularity of the classification; k represents the number of categories in the current live platform; ISiRepresenting a first similarity; TS represents a second similarity; DS (direct sequence)iIndicating a third degree of similarity.
8. The user classification system according to claim 7, wherein the predetermined similarity algorithm is specifically formulated as:
wherein s (u, v) represents the similarity between the user u and the user v, τ (u) represents the set of live broadcasting rooms watched by the user u, τ (v) represents the set of live broadcasting rooms watched by the user v, τ (u) # τ (v) represents the set of live broadcasting rooms watched by the user u and the user v together, w (u, x) represents the characteristic value of the live broadcasting room x watched by the user u in a preset time period, and w (v, x) represents the characteristic value of the live broadcasting room x watched by the user v in the preset time period;
the characteristic value is obtained by acquiring the behavior characteristics of the user in the corresponding live broadcast room and scoring according to preset behavior evaluation indexes.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810739572.XA CN110689035B (en) | 2018-07-06 | 2018-07-06 | User classification method, storage medium, electronic device and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810739572.XA CN110689035B (en) | 2018-07-06 | 2018-07-06 | User classification method, storage medium, electronic device and system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110689035A CN110689035A (en) | 2020-01-14 |
| CN110689035B true CN110689035B (en) | 2022-06-24 |
Family
ID=69107108
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810739572.XA Active CN110689035B (en) | 2018-07-06 | 2018-07-06 | User classification method, storage medium, electronic device and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110689035B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111263179A (en) * | 2020-02-20 | 2020-06-09 | 广州虎牙科技有限公司 | Live broadcast room drainage method, device, system, server and storage medium |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105869034A (en) * | 2016-04-06 | 2016-08-17 | 无锡中科富农物联科技有限公司 | Agricultural material consumption data clustering method based on improved Chameleon algorithm |
| CN106127519A (en) * | 2016-06-24 | 2016-11-16 | 武汉斗鱼网络科技有限公司 | A kind of live platform user divided method based on K Means algorithm and system |
| CN106780064A (en) * | 2016-12-02 | 2017-05-31 | 腾讯科技（深圳）有限公司 | A kind of region partitioning method, device and the network equipment |
-
2018
- 2018-07-06 CN CN201810739572.XA patent/CN110689035B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105869034A (en) * | 2016-04-06 | 2016-08-17 | 无锡中科富农物联科技有限公司 | Agricultural material consumption data clustering method based on improved Chameleon algorithm |
| CN106127519A (en) * | 2016-06-24 | 2016-11-16 | 武汉斗鱼网络科技有限公司 | A kind of live platform user divided method based on K Means algorithm and system |
| CN106780064A (en) * | 2016-12-02 | 2017-05-31 | 腾讯科技（深圳）有限公司 | A kind of region partitioning method, device and the network equipment |
Non-Patent Citations (2)
| Title |
|---|
| 基于相似性模块度最大约束标记传播的网络社团发现算法;黄健斌等;《北京大学学报(自然科学版)》;20130531;第49卷(第03期);第389-396页 * |
| 基于节点相似性度量的社团结构划分方法;梁宗文等;《计算机应用》;20150510;第35卷(第05期);第1213-1223页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110689035A (en) | 2020-01-14 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US20190205704A1 (en) | Method for Training Model and Information Recommendation System | |
| CN108090208A (en) | Fused data processing method and processing device | |
| CN103500212B (en) | A kind of multimedia file recommendation method and electronic equipment | |
| CN107864405B (en) | Viewing behavior type prediction method, device and computer readable medium | |
| CN111209477B (en) | Information recommendation method and device, electronic equipment and storage medium | |
| CN111026969B (en) | Content recommendation method and device, storage medium and server | |
| CN111523035B (en) | Recommendation method, device, server and medium for APP browsing content | |
| US20160314484A1 (en) | Method and system for mining churn factor causing user churn for network application | |
| US10248864B2 (en) | Systems and methods for contextual video shot aggregation | |
| CN108595526A (en) | Resource recommendation method and device | |
| CN110689035B (en) | User classification method, storage medium, electronic device and system | |
| CN104679791A (en) | Processing method and device for acquiring data packets | |
| CN113301017B (en) | Attack detection and defense method and device based on federal learning and storage medium | |
| CN109660831B (en) | Barrage sending period setting method, storage medium, electronic device and system | |
| CN104506895A (en) | Video classification method | |
| CN114449342B (en) | Video recommendation method, device, computer readable storage medium and computer equipment | |
| CN113254919B (en) | Abnormal device identification method, electronic device, and computer-readable storage medium | |
| CN117272056A (en) | Object feature construction method, device and computer readable storage medium | |
| CN106610991A (en) | Data processing method and device | |
| Smaaberg et al. | A User-Study on Context-aware Group Recommendation for Concerts. | |
| CN114969493A (en) | Content recommendation method and related device | |
| CN114154014A (en) | Video cold start recommendation method and device | |
| CN114118651A (en) | Evaluation method, device, equipment and computer storage medium | |
| Sunitha et al. | Addressing long tail problem in music recommendation systems | |
| CN117112870B (en) | User interaction image classification method and AI session interaction system based on artificial intelligence |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| TR01 | Transfer of patent right | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20231213 Address after: Room 21032, Floor 9, Building 2, Micro World, No. 23, Century Avenue, Liuwu New District, Lhasa, Xizang Autonomous Region, 850000 Patentee after: Lhasa Jinlingteng Electronic Technology Co.,Ltd. Address before: 430000 East Lake Development Zone, Wuhan City, Hubei Province, No. 1 Software Park East Road 4.1 Phase B1 Building 11 Building Patentee before: WUHAN DOUYU NETWORK TECHNOLOGY Co.,Ltd. |