CN107730271A - Similar users based on virtual interacting object determine method, apparatus and electronic equipment - Google Patents
Similar users based on virtual interacting object determine method, apparatus and electronic equipment Download PDFInfo
- Publication number
- CN107730271A CN107730271A CN201710850770.9A CN201710850770A CN107730271A CN 107730271 A CN107730271 A CN 107730271A CN 201710850770 A CN201710850770 A CN 201710850770A CN 107730271 A CN107730271 A CN 107730271A
- Authority
- CN
- China
- Prior art keywords
- user
- grouping
- preference
- subgroup
- similarity
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Classifications
-
- 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
- G06Q30/01—Customer relationship services
-
- 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
Abstract
The embodiments of the invention provide the similar users based on virtual interacting object to determine method, apparatus and electronic equipment, is related to technical field of information recommendation, and this method includes:Obtain user grouping corresponding to each user mutual object;Each user grouping is divided into by multiple subgroups by default rule of classification, wherein, for each user grouping, the quantity of user is respectively less than the quantity of user in one user grouping in each subgroup of a user grouping;Respectively each subgroup builds and associates virtual interacting object, obtains virtual objects user grouping corresponding to multiple virtual interacting objects;Calculate the similarity between the user in each virtual objects user grouping;Similarity with targeted customer is met to similar users of the user as the targeted customer of default condition of similarity.The similar users based on virtual interacting object of the embodiment of the present invention determine that method can reduce the computation complexity for determining similar users, facilitate practical application.
Description
Technical field
The present invention relates to technical field of information recommendation, more particularly to the similar users determination side based on virtual interacting object
Method, device and electronic equipment.
Background technology
Collaborative filtering can excavate the potential characteristic of user and article by the various operation informations of user, so as to
User and user are further discovered that, associating between article and article, and user and article.The existing recommendation based on collaborative filtering
Algorithm is broadly divided into three classes:Collaborative filtering based on user, the collaborative filtering based on article and the collaborative filtering based on model.
During information recommendation, in order to more accurately recommend the service that it admires to user, the proposed algorithm based on collaborative filtering is extensive
Applied in various commending systems.For example, Netflix and YouTube are recommended using based on the collaborative filtering of model to user
Video service, Amazon are used and once employed based on user's to user's Recommendations, Yahoo based on the collaborative filtering of article
Collaborative filtering recommends news to user.
The existing collaborative filtering method based on user, it is thus necessary to determine that similar to the hobby of targeted customer or taste is similar
User:Preference of each user to each actuals is obtained, then by providing algorithm, using each user to each actual business
The preference of product calculates the similarity between all users, selection and targeted customer similarity highest user, as target
The similar users of user.
However, inventor has found that at least there are the following problems for prior art during the present invention is realized:
When number of users is larger, in order to determine the similar users of targeted customer, each user is being utilized to each actuals
Preference when calculating the similarity between all users, it is necessary to the combination of the user calculated is very more, determine similar users
Computation complexity it is higher, be difficult to apply in practice.
The content of the invention
The purpose of the embodiment of the present invention is that providing a kind of similar users based on virtual interacting object determines method, apparatus
And electronic equipment, the computation complexity for determining similar users is reduced to realize, facilitates practical application.Concrete technical scheme is as follows:
In a first aspect, the embodiments of the invention provide a kind of similar users based on virtual interacting object to determine method, institute
The method of stating includes:
User grouping corresponding to each user mutual object is obtained, wherein, each user grouping includes and a use
The preference of interactive each user and each user to one user mutual object be present in family interactive object;
Each user grouping is divided into by multiple subgroups by default rule of classification, wherein, for each use
Family is grouped, and the quantity of user is respectively less than the quantity of user in one user grouping in each subgroup of a user grouping;
Respectively each subgroup builds and associates virtual interacting object, obtains corresponding to multiple virtual interacting objects
Virtual objects user grouping, wherein, for each subgroup, each user is to one subgroup pair in a subgroup
The preference for the virtual interacting object answered, it is corresponding to one subgroup with each user in one subgroup respectively
The preference of user mutual object corresponding to user grouping is identical;
Calculate the similarity between the user in each virtual objects user grouping;
Similarity with targeted customer is met to similar users of the user as the targeted customer of default condition of similarity.
With reference in a first aspect, in first aspect in the first possible embodiment, each user mutual is obtained described
Before user grouping corresponding to object, methods described also includes:
Preference of each user to each user mutual object is obtained respectively;
It is divided into one group, and preference corresponding to association by all users interacted be present with each user mutual object,
Obtain user grouping corresponding to each user mutual object.
With reference to first aspect or first aspect the first possible embodiment, in second of possible implementation of first aspect
It is described that each user grouping is divided into by multiple subgroups by default rule of classification in mode, including:
For each user grouping, by the preference descending of user to user interactive object in the user grouping
Arrangement, obtains user preference degree of sequence corresponding to each user grouping;
For each preference sequence, will with corresponding user in the preference sequence, according to it is described partially
The order of good degree of sequence is ranked up, and obtains user's sequence corresponding to each user grouping;
In each user's sequence, according to the sequence of user, pre-set user quantity user is chosen successively and is divided into
One group, obtain multiple subgroups.
With reference to second of possible embodiment of first aspect, in first aspect in the third possible embodiment, institute
State in each user's sequence, according to the sequence of user, choose pre-set user quantity user successively and be divided into one group, obtain
To multiple subgroups, including:
In each user's sequence, according to the sequence of user, pre-set user quantity user is chosen successively and is divided into
One group, when the quantity of remaining ungrouped user is less than the pre-set user quantity, remaining ungrouped user is divided into
One group, obtain multiple subgroups.
With reference to first aspect or first aspect the first possible embodiment, in the 4th kind of possible implementation of first aspect
In mode, the similarity calculated in each virtual objects user grouping between user, including:
Similarity in each virtual objects user grouping of parallel computation between user.
With reference to first aspect or first aspect the first possible embodiment, in the 5th kind of possible implementation of first aspect
In mode, meet the user of default condition of similarity as the similar of the targeted customer in the similarity using to targeted customer
After user, methods described also includes:
The preference for choosing the similar users is more than default preference threshold value, and the targeted customer does not interact
The user mutual object crossed, is pushed to the targeted customer.
Second aspect, the embodiments of the invention provide a kind of similar users determining device based on virtual interacting object, institute
Stating device includes:
User grouping acquisition module, for obtaining user grouping corresponding to each user mutual object, wherein, it is each described
User grouping includes each user interacted and each user being present to one user mutual pair with a user mutual object
The preference of elephant;
Subgroup division module, for each user grouping to be divided into multiple subgroups by default rule of classification,
Wherein, for each user grouping, the quantity of user is respectively less than one use in each subgroup of a user grouping
The quantity of user in the packet of family;
Virtual interacting object relating module, built for respectively each subgroup and associate virtual interacting object,
Virtual objects user grouping corresponding to multiple virtual interacting objects is obtained, wherein, for each subgroup, a subgroup
In each user to the preference of virtual interacting object corresponding to one subgroup, respectively with it is each in one subgroup
User is identical to the preference of the user mutual object corresponding to user grouping corresponding to one subgroup;
Similarity calculation module, for calculating the similarity between the user in each virtual objects user grouping;
Similar users choose module, for the similarity with targeted customer to be met to the user of default condition of similarity as institute
State the similar users of targeted customer.
It is described based on virtual interacting object in second aspect in the first possible embodiment with reference to second aspect
Similar users determining device also includes:
Preference acquisition module, for obtaining preference of each user to each user mutual object respectively;
User grouping division module, it is divided into one group for all users interacted will be present with each user mutual object,
And preference corresponding to associating, obtain user grouping corresponding to each user mutual object.
With reference to second aspect or second aspect the first possible embodiment, in second of possible implementation of second aspect
In mode, the subgroup division module, including:
Preference sorting sub-module, for for each user grouping, by user in the user grouping to
The preference descending arrangement of family interactive object, obtains user preference degree of sequence corresponding to each user grouping;
User's sorting sub-module, for for each preference sequence, will with it is right in the preference sequence
The user answered, it is ranked up according to the order of the preference sequence, obtains user's sequence corresponding to each user grouping
Row;
User divides submodule, in each user's sequence, according to the sequence of user, choosing default use successively
Amount amount user is divided into one group, obtains multiple subgroups.
With reference to second of possible embodiment of second aspect, in second aspect in the third possible embodiment, institute
State user and divide submodule, be specifically used for:
In each user's sequence, according to the sequence of user, pre-set user quantity user is chosen successively and is divided into
One group, when the quantity of remaining ungrouped user is less than the pre-set user quantity, remaining ungrouped user is divided into
One group, obtain multiple subgroups.
With reference to second aspect or second aspect the first possible embodiment, in the 4th kind of possible implementation of second aspect
In mode, the similarity calculation module, it is specifically used for:
Similarity in each virtual objects user grouping of parallel computation between user.
With reference to second aspect or second aspect the first possible embodiment, in the 5th kind of possible implementation of second aspect
In mode, the similar users determining device based on virtual interacting object also includes:
Pushing module, the preference for choosing the similar users are more than default preference threshold value, and the mesh
The user mutual object that mark user did not interact, is pushed to the targeted customer.
The third aspect, the embodiments of the invention provide a kind of electronic equipment, including processor, communication interface, memory and
Communication bus, wherein, the processor, the communication interface, the memory complete mutual communication by communication bus;
The memory, for depositing computer program;
The processor, during for performing the program deposited on the memory, realize above-mentioned first aspect and first
The first any described method and step into the 5th kind of possible embodiment of aspect.
Fourth aspect, the embodiments of the invention provide a kind of computer-readable recording medium, the computer-readable storage
Instruction is stored with medium, when run on a computer so that computer performs above-mentioned first aspect and first aspect the
A kind of either method in the 5th kind possible embodiment.
5th aspect, the embodiments of the invention provide a kind of computer program product for including instruction, when it is in computer
During upper operation so that computer perform above-mentioned first aspect and first aspect the first into the 5th kind of possible embodiment
Either method.
Similar users provided in an embodiment of the present invention based on virtual interacting object determine method, apparatus and electronic equipment,
This method includes:Obtain user grouping corresponding to each user mutual object;Will each user by default rule of classification
It is divided into multiple subgroups, wherein, for each user grouping, the number of user in each subgroup of a user grouping
Amount is respectively less than the quantity of user in one user grouping;Respectively each subgroup builds and associates virtual interacting pair
As obtaining virtual objects user grouping corresponding to multiple virtual interacting objects;Calculate the user in each virtual objects user grouping
Between similarity;Similarity to targeted customer is met to similar use of the user as the targeted customer of default condition of similarity
Family.In the embodiment of the present invention, by the way that user is divided into multigroup virtual objects user grouping, and used in virtual objects user grouping
The quantity at family is less than the quantity of user in user grouping, calculates the similarity between the user in each virtual objects user grouping, and
The similar users of targeted customer are chosen by presetting condition of similarity, the computation complexity for determining similar users can be reduced, it is convenient
Practical application.Certainly, all advantages for reaching the above simultaneously must be not necessarily required to by implementing any product of the present invention or method.
Brief description of the drawings
In order to illustrate more clearly about the embodiment of the present invention or technical scheme of the prior art, below will be to embodiment or existing
There is the required accompanying drawing used in technology description to be briefly described.
Fig. 1 is that the similar users based on virtual interacting object of the embodiment of the present invention determine the schematic flow sheet of method;
Fig. 2 is the schematic flow sheet of the method for determination user's similarity of the embodiment of the present invention;
Fig. 3 is the schematic diagram of the similar users determining device based on virtual interacting object of the embodiment of the present invention;
Fig. 4 is the schematic diagram of the electronic equipment of the embodiment of the present invention.
Embodiment
Below in conjunction with the accompanying drawing in the embodiment of the present invention, the technical scheme in the embodiment of the present invention is described.
The existing collaborative filtering method based on user, it is determined that during similar users, it is necessary to calculate all users between
Similarity, then according to the similarity between all users, by default selection rule, choose the phase of the targeted customer specified
Like user.The existing method for determining similar users is larger in number of users, it is necessary to calculate the similarity between all users
When, computation complexity is high, is difficult to apply in reality.In view of the above-mentioned problems, the embodiment of the present invention based on virtual interacting object
Similar users determine method, apparatus and electronic equipment, by the way that user is divided into multigroup virtual objects user grouping, and it is virtual right
As user in user grouping quantity be less than user grouping in user quantity, calculate the user in each virtual objects user grouping
Between similarity, and by preset condition of similarity choose targeted customer similar users, can reduce determine similar users meter
Complexity is calculated, facilitates practical application.
Referring to Fig. 1, present invention implementation provides a kind of similar users based on virtual interacting object and determines method, this method
Including:
S101, user grouping corresponding to each user mutual object is obtained, wherein, each user grouping includes and a use
The preference of interactive each user and each user to a user mutual object be present in family interactive object.
For example, it is that any user in all user groupings is grouped to make the first user grouping, the first user grouping pair
The user mutual object answered is the first user mutual object, then the first user grouping includes:Deposited with the first user mutual object
Preference journey of each user interacted to the first user mutual object be present in interactive each user, and with the first user mutual object
Degree.
User mutual object is any object for meeting the embodiment of the present invention, including actual product and Virtual Service.For example,
Picture, music, game, video, food, ornaments, household services and tourist service etc..
S102, each user grouping is divided into by multiple subgroups by default rule of classification, wherein, for each user point
Group, the quantity of user is respectively less than the quantity of user in a user grouping in each subgroup of a user grouping.
Default rule of classification is to meet any rule of classification that the present invention is implemented, for example, in a user grouping every time
Choose pre-set user quantity user and be divided into a subgroup, until all user's divisions finish in the user grouping.Or, will
Preference is divided into several gradients from minimum to maximum, and by the preference and preference pair in a user grouping
The user answered is divided into corresponding gradient, obtains multiple subgroups.
S103, respectively each subgroup build and associate virtual interacting object, and it is corresponding to obtain multiple virtual interacting objects
Virtual objects user grouping, wherein, for each subgroup, each user is to empty corresponding to a subgroup in a subgroup
Intend the preference of interactive object, respectively with each user in a subgroup to corresponding to the corresponding user grouping of a subgroup
User mutual object preference it is identical.
For example, the first user grouping (the first user grouping is any user packet in all user groupings) is corresponding
User mutual object be the first user mutual object, (the first virtual objects user grouping is the first virtual objects user grouping
Virtual objects user grouping corresponding to any subgroup of first user grouping division) corresponding to virtual interacting object be first empty
Intend interactive object, then the first user (the first user is any user in the first virtual objects user grouping) is virtual to first hands over
The preference of mutual object, identical to the preference of the first user mutual object with the first user, i.e., the first user is first
User preference degree in virtual objects user grouping, with user preference degree phase of first user in the first user grouping
Together.
S104, calculate the similarity between the user in each virtual objects user grouping.
For each virtual objects user grouping in each virtual objects user grouping, calculate in virtual objects user grouping
Similarity between user;Using the second virtual objects user grouping, (the second virtual objects user grouping is each virtual objects user point
Any virtual objects user grouping in group) exemplified by, obtain each user in the second virtual objects user grouping and all users are handed over
The preference of mutual object, and by default similarity calculating method, calculate each user in the second virtual objects user grouping
Similarity.Default similarity calculating method is any means for meeting the embodiment of the present invention herein, as included angle cosine is similar
Computational methods are spent, manhatton distance calculates similarity based method, and Euclidean distance calculates similarity based method.
The similarity between the user in each virtual objects user grouping is calculated in embodiments of the present invention, and virtual objects are used
Number of users in the packet of family is less than the quantity of user in corresponding user grouping, compared to user in each user grouping of calculating
Quantity, the complexity of calculating can be reduced.
S105, the similarity to targeted customer is met to similar use of the user as targeted customer of default condition of similarity
Family.
Default similarity condition herein is to meet any condition of the embodiment of the present invention, can be carried out according to actual conditions
Setting, for example, by with the similarity of targeted customer be more than similarity threshold (similarity threshold is set according to actual conditions,
Such as 60% or higher, 70% or higher, or 80% or higher) similar users of the user as targeted customer;Or, will be with target
Similar users of the similarity highest user of user as targeted customer.According to the size of the similarity to targeted customer, choosing
Take that M's before accounting (M is set according to actual conditions, such as 20% or smaller, 10% or smaller, or 5% or smaller) is most like
User as similar users (according to the number of the M similar users calculated to round up, such as the phase calculated according to M
Quantity like user is 5.1, then actual to choose before similarity ranking 66 users as similar users).Choose and used with target
Similar users of the similarity high user in family as targeted customer, determine that the accuracy of similar users is high.
In embodiments of the present invention, user grouping is divided into multigroup virtual objects user grouping, and virtual objects user
The quantity of user is less than the quantity of user in user grouping in packet, calculates the phase between the user in each virtual objects user grouping
Like degree, and by presetting the similar users of condition of similarity selection targeted customer, determine the computation complexity of the method for similar users
It is low, facilitate practical application.
Optionally, with reference to above-described embodiment, in another embodiment provided by the invention, passing through in above-mentioned steps S102
Each user grouping is divided into multiple subgroups by default rule of classification, may comprise steps of:
Step 1, for each user grouping, by the preference descending of user to user interactive object in user grouping
Arrangement, obtains user preference degree of sequence corresponding to each user grouping.
Step 2, for each preference sequence, will user corresponding with preference sequence, according to preference
The order of sequence is ranked up, and obtains user's sequence corresponding to each user grouping.
Step 3, in each user's sequence, according to the sequence of user, pre-set user quantity user division is chosen successively
For one group, multiple subgroups are obtained.
Certainly, except comprising with open air, further comprises user to user mutual object corresponding to the subgroup in subgroup
Preference.Pre-set user quantity is to meet any number of the embodiment of the present invention, according to the actual computing capability of processor
It is determined.Such as two or more is could be arranged to, 5 or bigger, 50 or bigger, 100 or bigger etc..The value of pre-set user quantity
What is set is smaller, determines that the calculating speed of similar users is faster.In addition, for different user groupings, can set different
Pre-set user quantity, can be that pre-set user quantity sets larger value when the user included in some user grouping is less.
Each user grouping is directed in embodiments of the present invention, preference can also be arranged using descending, obtained partially
Good degree descending sequence, the user in each user grouping is arranged according to the order of preference descending sequence accordingly
Sequence, and choose pre-set user quantity user successively and be divided into one group, obtain multiple subgroups.
In embodiments of the present invention, each user in same subgroup is to the inclined of user mutual object corresponding to the subgroup
Good degree is close, and the hobby of each user in same subgroup is also just more close, and the similar users accuracy subsequently determined is high.
Optionally, in above-described embodiment step 3 in each user's sequence, according to the sequence of user, choose successively pre-
If number of users user is divided into one group, multiple subgroups are obtained, can also specifically be included:
In each user's sequence, according to the sequence of user, pre-set user quantity user is chosen successively and is divided into one group,
When the quantity of remaining ungrouped user is less than pre-set user quantity, remaining ungrouped user is divided into one group, obtained
Multiple subgroups.
It is ranked into for second user packet (any user that second user is grouped into all user groupings is grouped)
Second user sequence, after N (N is positive integer) individual subgroup is marked off, when dividing N+1 group subgroups, if now remaining
The quantity of user be less than pre-set user quantity, remaining user is divided into N+1 groups.For example, it is in pre-set user quantity
5, after marking off N number of subgroup, remaining user is user 1 (corresponding preference 1), user 2 (corresponding preference 2), is used
Family 3 (corresponding preference 3), then N+1 groups subgroup is:User 1, and preference 1, user 2, and preference 2, user 3, partially
Good degree 3.
In embodiments of the present invention, the subgroup division side when remaining users quantity is less than pre-set user quantity is given
Method, subgroup division is reasonable, and the similar users accuracy subsequently determined is high.
Optionally, in another possible embodiment provided in an embodiment of the present invention, the step in above-described embodiment
Three kinds in each user's sequence, according to the sequence of user, choose pre-set user quantity user successively and be divided into one group, obtain
To multiple subgroups, can also include:
In each user's sequence, according to the sequence of user, pre-set user quantity user is chosen successively and is divided into one group,
When the quantity of remaining ungrouped user is less than pre-set user quantity, the user of respective numbers is chosen, will as duplicate customer
Duplicate customer is divided into one group with remaining ungrouped user, obtains multiple subgroups, wherein, the number of remaining ungrouped user
Amount and duplicate customer quantity and be pre-set user quantity.
It is ranked into for second user packet (any user that second user is grouped into all user groupings is grouped)
Second user sequence, after N (N is positive integer) individual subgroup is marked off, when dividing N+1 group subgroups, if now remaining
The quantity of user be less than pre-set user quantity, the user of respective numbers can be randomly selected in n-th subgroup, as weight
It is multiplexed family.Duplicate customer and remaining user quantity and, be pre-set user quantity.Duplicate customer and remaining user are drawn
It is divided into N+1 group subgroups.For example, when pre-set user quantity is 3, n-th subgroup is:User 1, preference 1,
User 2, and preference 2, user 3, preference 3.Remaining users are user 4, preference 4.Then N+1 groups subgroup can be with
For:User 2, and preference 2, user 3, and preference 3, user 4, preference 4.
In embodiments of the present invention, the subgroup division side when remaining users quantity is less than pre-set user quantity is given
Method, subgroup division is reasonable, and the similar users accuracy subsequently determined is high.
Optionally, in another possible embodiment provided in an embodiment of the present invention, the step in above-described embodiment
Three kinds in each user's sequence, according to the sequence of user, choose pre-set user quantity user successively and be divided into one group, obtain
To multiple subgroups, can also include:
For each user's sequence, if the quantity of user is less than default overlapping startup threshold value in same user's sequence,
According to the sequence of user, eclipsed form chooses pre-set user quantity user and is divided into one group successively, obtains multiple subgroups.
Default overlapping startup threshold value is to meet any number of the embodiment of the present invention, such as 5 or bigger, 50 or bigger, 100
It is or bigger etc..The number of users of each overlapping selection is set according to actual conditions, but should be less than pre-set user quantity.
For example, when user's sequence is user 1, user 2, user 3, user 4, user 5, user 6, user 7, user 8 and user 9, in advance
If number of users is 5, the default overlapping threshold value that starts is 10, every time 3 users of overlapping selection, then subgroup is (user 1, use
Family 2, user 3, user 4, user 5), (user 3, user 4, user 5, user 6, user 7) and (user 5, user 6, user 7, use
Family 8, user 9).In embodiments of the present invention, when dividing subgroup, appropriate overlapping user, can obtain between more users
Similarity.
Optionally, before user grouping corresponding to each user mutual object is obtained, this method also includes:
Step 1, preference of each user to each user mutual object is obtained respectively.
Optionally, before each user is obtained respectively to the preference of each user mutual object, based on virtual interacting pair
The similar users of elephant determine that method also includes:Interaction scenario of the user to each user mutual object is obtained, it is determined that each user
Preference in special time to each user mutual object.For example, by viewing duration divided by video of certain user to a video
Total duration, the preference as the user to the video;Or, by scoring of the user to a user mutual object divided by completely
Point, the preference as the user to the user mutual object.In embodiments of the present invention, according to user and interactive object
Interaction scenario determines preference, and preference determines accurate.
Step 2, it is divided into one group by all users interacted be present with each user mutual object, and corresponding to association partially
Good degree, obtain user grouping corresponding to each user mutual object.
Corresponding preference refers to that user is to user mutual object corresponding to the user grouping in a user grouping
Preference.
Optionally, in a kind of possible embodiment of the embodiment of the present invention, when number of users is very big, count for convenience
Calculate, can only choose compared with user corresponding to the user mutual object of unexpected winner, and compared with user corresponding to the user mutual object of unexpected winner
User grouping is formed to the preference of the user mutual object compared with unexpected winner.Wherein, the user mutual object compared with unexpected winner is generation
Interactive number of users is less than default interaction threshold value, and (default interaction threshold value is carried out according to the total quantity and the actual demand that calculates of user
The 50% or smaller of setting, such as total number of users, the 20% or smaller of total number of users, or the 5% or smaller of total number of users) user
Interactive object.It is the use interacted be present with the user mutual object compared with unexpected winner compared with user corresponding to the user mutual object of unexpected winner
Family.The user mutual object compared with unexpected winner is chosen, obtains the user grouping of the user mutual object compared with unexpected winner, user in user grouping
Quantity it is smaller, it is follow-up determine similar users computation complexity it is low.
In embodiments of the present invention, all users interacted will be present with same user mutual object and be divided into a user
In packet, user in same user grouping has the association for same user mutual object, subsequently to calculate between each user
Similarity provide premise.
Optionally, with reference to above-described embodiment, in an embodiment provided by the invention, the calculating in above-mentioned steps S104 is each
Similarity in virtual objects user grouping between user, it can include:
Similarity in each virtual objects user grouping of parallel computation between user.
In the similarity between calculating user, because each virtual objects user grouping each corresponds to corresponding virtual interacting pair
As, therefore each virtual interacting object can implement parallel computation, calculating speed is fast.Used compared to directly calculating in each user grouping
Similarity between family, the complexity of calculating is not only reduced, the parallel meter of similarity between user in user grouping can also be realized
Calculate, Similarity Measure speed is fast, efficiency high.
Optionally, the similarity to targeted customer is being met to the user of default condition of similarity as the similar of targeted customer
After user, this method also includes:
The preference for choosing similar users is more than default preference threshold value, and the user that targeted customer did not interact
Interactive object, it is pushed to targeted customer.
Default preference threshold value herein is to meet any number of the embodiment of the present invention, for example, 70% or higher,
80% or higher, or 90% or higher.User mutual object is pushed for targeted customer herein, can apply and meet the present invention in fact
It is such as user's pushing video in any scene for applying example, pushes news for user, played to be pushed for user, and pushed away for user
Send individualized homepage etc..
In embodiments of the present invention, the preference of similar users is more than default preference threshold value, and targeted customer
The user mutual object not interacted, is pushed to targeted customer, and the user mutual object of push more conforms to targeted customer's
Interest, the degree of accuracy of user mutual Object Push are high.
Referring to Fig. 2, Fig. 2 be the embodiment of the present invention determination user between similarity method schematic flow sheet, including:
S201, user and the interaction scenario of each user mutual object are obtained, by default preference algorithm, obtain each use
Family is in special time to the preference of corresponding user mutual object.
After user interacts with the generation of user mutual object, the interactive information of each user and each user mutual object are collected,
And handled, to obtain each user in special time to the preference of each user mutual object.For example, obtain user couple
The viewing duration of certain video, the business that the total duration for watching duration divided by video is obtained, the preference journey as user to the video
Degree.User may also differ in different time to user to the preference of interactive object, such as when temperature is higher, user
Also can be higher to the preference of cold drink;At night, the video of terrified suspense class is more by user preference.By user to user interaction pair
The preference association special time of elephant, the preference accuracy of user to user interactive object are high.
S202, the corresponding relation of each user and each interactive object of user are established, to determine user grouping.
Preference according to each user to each user mutual object, it is corresponding with each interactive object of user to establish each user
Relation, to determine user grouping.The corresponding user mutual object of one user grouping, user grouping are included to the user point
Each user of interaction, and corresponding preference be present in user mutual object corresponding to group.Such as establish user-user interaction pair
As form, as shown in table 1, wherein, "" represent between user and user mutual object be not present interact.
Table 1
For example, in user grouping corresponding to user mutual object D, preference journey of the user 1 to user mutual object D
Spend for 0.3, user 3 is 0.6 to user mutual object D preference, and user 4 is to user mutual object D preference
0.7, user 5 is 0.1 to user mutual object D preference, and user 6 is 0.2 to user mutual object D preference.
S203, for each user grouping, the preference according to each user to corresponding user mutual object, by pre-
If ordering rule, each user is ranked up, obtains multiple user's sequences.
Based on each user mutual object, interactive users to the user mutual object be present by all, according to user couple
The descending sort of the user mutual object preference, is ranked up to user accordingly, obtains multiple user's sequences.With table 1
Exemplified by, for user mutual object A, user's sequence is:1,5,4,3,6;For user mutual object E, user's sequence is:2,4,
6,5,1,3.
S204, according to multiple user's sequences, by default rule of classification, the user in user grouping is grouped,
Obtain multiple subgroups.
Number of users in each subgroup being set, being divided each user interacted be present with same user mutual object
Group.By taking table 1 as an example, when number of users in subgroup is 3, for user mutual object A, obtain two subgroups (1,5,
And (3,6) 4);For user mutual object E, two subgroups (2,4,6) and (5,3,1) are obtained.For user mutual A
Packet and user mutual object E subgroup, can both obtain the similarity of user 5 and user 4, can obtain again user 5 and
The similarity of user 3, i.e., same user may be present in different subgroups.And the actual cosine phase of user 5 and user 4
Like similar in the actual cosine similarity of degree and user 5 and user 3 is comparison, therefore the division of the subgroup of the embodiment of the present invention
Method is still relatively more rational.
S205, build for each subgroup and associate a virtual interacting object, obtain multiple virtual objects user groupings,
Wherein, all virtual interacting objects differ.
By taking table 1 as an example, when the number of users in subgroup is 3, for user mutual object E, two subgroups are obtained
(2,4,6) and (5,3,1).Built to subgroup (2,4,6) and associate a virtual interacting object VE1, give subgroup (5,3,1)
Build and associate a virtual interacting object VE2, two virtual objects user groupings are obtained, as shown in table 2.
Table 2
S206, the similarity in each virtual objects user grouping of parallel computation between user.
In each virtual objects user grouping, the similarity in calculating group between each user (can be according to user to all
The preference of user mutual object, the similarity between user is calculated by Jaccard formula or cosine similarity).It is because every
The corresponding unique virtual interacting object of individual virtual objects user grouping, therefore each virtual objects user grouping can be implemented simultaneously
Row calculates.By taking table 2 as an example, while calculate virtual interacting object VE1Corresponding virtual objects user grouping and virtual interacting object
VE2In corresponding virtual objects user grouping, the similarity between each user.
In embodiments of the present invention, user is ranked up according to preference and is divided into subgroup, respectively used in subgroup
Family is close to the preference of same user mutual object, and the hobby of each user is also just more nearly in same subgroup, calculates
The accuracy of the similarity gone out is high.Number of the number of users less than user in original subscriber's packet in virtual objects user grouping simultaneously
Amount, computation complexity is low, and can preferably implement parallel computation, Similarity Measure efficiency high.
Referring to Fig. 3, Fig. 3 is the similar users determining device based on virtual interacting object of the embodiment of the present invention, the device
Including:
User grouping acquisition module 301, for obtaining user grouping corresponding to each user mutual object, wherein, each
User grouping includes each user interacted and each user being present to user mutual object with user mutual object
Preference;
Subgroup division module 302, for each user grouping to be divided into multiple subgroups by default rule of classification, its
In, for each user grouping, the quantity of user is respectively less than in a user grouping and used in each subgroup of a user grouping
The quantity at family;
Virtual interacting object relating module 303, for being respectively that each subgroup builds and associates virtual interacting object, obtain
To virtual objects user grouping corresponding to multiple virtual interacting objects, wherein, for each subgroup, respectively used in a subgroup
Family is divided a son the preference of virtual interacting object corresponding to a subgroup with each user in a subgroup respectively
The preference of user mutual object corresponding to user grouping corresponding to group is identical;
Similarity calculation module 304, for calculating the similarity between the user in each virtual objects user grouping;
Similar users choose module 305, and the user for the similarity with targeted customer to be met to default condition of similarity makees
For the similar users of targeted customer.
In embodiments of the present invention, user is divided into multigroup virtual objects user grouping, and virtual objects user grouping
The quantity of middle user is less than the quantity of user in user grouping, calculates similar between the user in each virtual objects user grouping
Degree, and by presetting the similar users of condition of similarity selection targeted customer, determine that the computation complexity of the method for similar users is low,
Facilitate practical application.
Optionally, the similar users determining device based on virtual interacting object of the embodiment of the present invention also includes:
Preference acquisition module, for obtaining preference of each user to each user mutual object respectively;
User grouping division module, it is divided into one group for all users interacted will be present with each user mutual object,
And preference corresponding to associating, obtain user grouping corresponding to each user mutual object.
In embodiments of the present invention, all users interacted will be present with same user mutual object and be divided into a user
In packet, user in same user grouping has the association for same user mutual object, subsequently to calculate between each user
Similarity provide premise.
Optionally, subgroup division module, including:
Preference sorting sub-module, for for each user grouping, by user to user interaction pair in user grouping
The preference descending arrangement of elephant, obtains user preference degree of sequence corresponding to each user grouping;
User's sorting sub-module, for for each preference sequence, will user corresponding with preference sequence,
Order according to preference sequence is ranked up, and obtains user's sequence corresponding to each user grouping;
User divides submodule, in each user's sequence, according to the sequence of user, choosing pre-set user number successively
Amount user is divided into one group, obtains multiple subgroups.
In embodiments of the present invention, each user in same subgroup is to the inclined of user mutual object corresponding to the subgroup
Good degree is close, and the hobby of each user in same subgroup is also just more close, and the similar users accuracy subsequently determined is high.
Optionally, user divides submodule, is specifically used for:
In each user's sequence, according to the sequence of user, pre-set user quantity user is chosen successively and is divided into one group,
When the quantity of remaining ungrouped user is less than pre-set user quantity, remaining ungrouped user is divided into one group, obtained
Multiple subgroups.
In embodiments of the present invention, the subgroup division side when remaining users quantity is less than pre-set user quantity is given
Method, subgroup division is reasonable, and the similar users accuracy subsequently determined is high.
Optionally, similarity calculation module, it is specifically used for:
Similarity in each virtual objects user grouping of parallel computation between user.
In the similarity between calculating user, because each virtual objects user grouping each corresponds to corresponding virtual interacting pair
As, therefore parallel computation can be implemented, calculating speed is fast.Compared to the similarity directly calculated in each user grouping between user,
The complexity of calculating is not only reduced, the parallel computation of similarity between user in user grouping, Similarity Measure can also be realized
Speed is fast, efficiency high.
Optionally, the similar users determining device based on virtual interacting object of the embodiment of the present invention also includes:
Pushing module, the preference for choosing similar users is more than default preference threshold value, and targeted customer does not have
There is the user mutual object interacted, be pushed to targeted customer.
In embodiments of the present invention, the preference of similar users is more than default preference threshold value, and targeted customer
The user mutual object not interacted, is pushed to targeted customer, and the user mutual object of push more conforms to targeted customer's
Interest, the degree of accuracy of user mutual Object Push are high.
The embodiment of the present invention additionally provides a kind of electronic equipment, as shown in figure 4, including processor 401, communication interface 402,
Memory 403 and communication bus 404, wherein, processor 401, communication interface 402, memory 403 is complete by communication bus 404
Into mutual communication,
Memory 403, for depositing computer program;
Processor 401, during for performing the program deposited on memory 403, realize following steps:
Step 1, user grouping corresponding to each user mutual object is obtained, wherein, each user grouping includes and one
The preference of interactive each user and each user to a user mutual object be present in user mutual object.
Step 2, each user grouping is divided into by multiple subgroups by default rule of classification, wherein, for each user
It is grouped, the quantity of user is respectively less than the quantity of user in a user grouping in each subgroup of a user grouping.
Step 3, respectively each subgroup build and associate virtual interacting object, obtain multiple virtual interacting objects pair
The virtual objects user grouping answered, wherein, for each subgroup, each user is to corresponding to a subgroup in a subgroup
The preference of virtual interacting object, it is right to a subgroup corresponding user grouping institute with each user in a subgroup respectively
The preference for the user mutual object answered is identical.
Step 4, calculate the similarity between the user in each virtual objects user grouping.
Step 5, the similarity to targeted customer is met to similar use of the user as targeted customer of default condition of similarity
Family.
In embodiments of the present invention, user is divided into multigroup virtual objects user grouping, and virtual objects user grouping
The quantity of middle user is less than the quantity of user in user grouping, calculates similar between the user in each virtual objects user grouping
Degree, and by presetting the similar users of condition of similarity selection targeted customer, determine that the computation complexity of the method for similar users is low,
Facilitate practical application.
Certainly, processor 401, during for performing the program deposited on memory 403, additionally it is possible to realize above-described embodiment
In any similar users based on virtual interacting object determine method.
The communication bus that above-mentioned electronic equipment is mentioned can be Peripheral Component Interconnect standard (Peripheral Component
Interconnect, abbreviation PCI) bus or EISA (Extended Industry Standard
Architecture, abbreviation EISA) bus etc..The communication bus can be divided into address bus, data/address bus, controlling bus etc..
For ease of representing, only represented in figure with a thick line, it is not intended that an only bus or a type of bus.
The communication that communication interface is used between above-mentioned electronic equipment and other equipment.
Memory can include random access memory (Random Access Memory, abbreviation RAM), can also include
Nonvolatile memory (non-volatile memory), for example, at least a magnetic disk storage.Optionally, memory may be used also
To be at least one storage device for being located remotely from aforementioned processor.
Above-mentioned processor can be general processor, including central processing unit (Central Processing Unit,
Abbreviation CPU), network processing unit (Network Processor, abbreviation NP) etc.;It can also be digital signal processor
(Digital Signal Processing, abbreviation DSP), application specific integrated circuit (Application Specific
Integrated Circuit, abbreviation ASIC), field programmable gate array (Field-Programmable Gate Array,
Abbreviation FPGA) either other PLDs, discrete gate or transistor logic, discrete hardware components.
In another embodiment provided by the invention, a kind of computer-readable recording medium is additionally provided, the computer can
Read to be stored with instruction in storage medium, when run on a computer so that computer performs any institute in above-described embodiment
The similar users based on virtual interacting object stated determine method.
In embodiments of the present invention, user is divided into multigroup virtual objects user grouping, and virtual objects user grouping
The quantity of middle user is less than the quantity of user in user grouping, calculates similar between the user in each virtual objects user grouping
Degree, and by presetting the similar users of condition of similarity selection targeted customer, determine that the computation complexity of the method for similar users is low,
Facilitate practical application.
In another embodiment provided by the invention, a kind of computer program product for including instruction is additionally provided, when it
When running on computers so that computer performs any described similar use based on virtual interacting object in above-described embodiment
Family determines method.
In embodiments of the present invention, user is divided into multigroup virtual objects user grouping, and virtual objects user grouping
The quantity of middle user is less than the quantity of user in user grouping, calculates similar between the user in each virtual objects user grouping
Degree, and by presetting the similar users of condition of similarity selection targeted customer, determine that the computation complexity of the method for similar users is low,
Facilitate practical application.
In the above-described embodiments, can come wholly or partly by software, hardware, firmware or its any combination real
It is existing.When implemented in software, can realize in the form of a computer program product whole or in part.The computer program
Product includes one or more computer instructions.When loading on computers and performing the computer program instructions, all or
Partly produce according to the flow or function described in the embodiment of the present invention.The computer can be all-purpose computer, special meter
Calculation machine, computer network or other programmable devices.The computer instruction can be stored in computer-readable recording medium
In, or the transmission from a computer-readable recording medium to another computer-readable recording medium, for example, the computer
Instruction can pass through wired (such as coaxial cable, optical fiber, numeral from a web-site, computer, server or data center
User's line (DSL)) or wireless (such as infrared, wireless, microwave etc.) mode to another web-site, computer, server or
Data center is transmitted.The computer-readable recording medium can be any usable medium that computer can access or
It is the data storage devices such as server, the data center integrated comprising one or more usable mediums.The usable medium can be with
It is magnetic medium, (for example, floppy disk, hard disk, tape), optical medium (for example, DVD) or semiconductor medium (such as solid state hard disc
Solid State Disk (SSD)) etc..
For the embodiment of device/electronic equipment/storage medium/computer program product, because it is substantially similar
In embodiment of the method, so description is fairly simple, the relevent part can refer to the partial explaination of embodiments of method.
It should be noted that herein, such as first and second or the like relational terms are used merely to a reality
Body or operation make a distinction with another entity or operation, and not necessarily require or imply and deposited between these entities or operation
In any this actual relation or order.Moreover, term " comprising ", "comprising" or its any other variant are intended to
Nonexcludability includes, so that process, method, article or equipment including a series of elements not only will including those
Element, but also the other element including being not expressly set out, or it is this process, method, article or equipment also to include
Intrinsic key element.In the absence of more restrictions, the key element limited by sentence "including a ...", it is not excluded that
Other identical element also be present in process, method, article or equipment including the key element.
Each embodiment in this specification is described by the way of related, identical similar portion between each embodiment
Divide mutually referring to what each embodiment stressed is the difference with other embodiment.It is real especially for system
For applying example, because it is substantially similar to embodiment of the method, so description is fairly simple, related part is referring to embodiment of the method
Part explanation.
The foregoing is merely illustrative of the preferred embodiments of the present invention, is not intended to limit the scope of the present invention.It is all
Any modification, equivalent substitution and improvements made within the spirit and principles in the present invention etc., are all contained in protection scope of the present invention
It is interior.
Claims (13)
1. a kind of similar users based on virtual interacting object determine method, it is characterised in that methods described includes:
User grouping corresponding to each user mutual object is obtained, wherein, each user grouping includes handing over a user
The preference of interactive each user and each user to one user mutual object be present in mutual object;
Each user grouping is divided into by multiple subgroups by default rule of classification, wherein, for each user point
Group, the quantity of user is respectively less than the quantity of user in one user grouping in each subgroup of a user grouping;
Respectively each subgroup builds and associates virtual interacting object, obtains virtual corresponding to multiple virtual interacting objects
Object user is grouped, wherein, for each subgroup, each user is to corresponding to one subgroup in a subgroup
The preference of virtual interacting object, respectively with each user in one subgroup to the corresponding user of one subgroup
The preference of the corresponding user mutual object of packet is identical;
Calculate the similarity between the user in each virtual objects user grouping;
Similarity with targeted customer is met to similar users of the user as the targeted customer of default condition of similarity.
2. according to the method for claim 1, it is characterised in that obtain user corresponding to each user mutual object described
Before packet, methods described also includes:
Preference of each user to each user mutual object is obtained respectively;
Being divided into one group, and preference corresponding to association by all users interacted be present with each user mutual object, obtaining
User grouping corresponding to each user mutual object.
3. method according to claim 1 or 2, it is characterised in that described by presetting rule of classification by each use
Family is divided into multiple subgroups, including:
For each user grouping, the preference descending of user to user interactive object in the user grouping is arranged
Row, obtain user preference degree of sequence corresponding to each user grouping;
For each preference sequence, will with corresponding user in the preference sequence, according to the preference journey
The order of degree series is ranked up, and obtains user's sequence corresponding to each user grouping;
In each user's sequence, according to the sequence of user, pre-set user quantity user is chosen successively and is divided into one group,
Obtain multiple subgroups.
4. according to the method for claim 3, it is characterised in that it is described in each user's sequence, according to user's
Sequence, pre-set user quantity user is chosen successively and is divided into one group, obtains multiple subgroups, including:
In each user's sequence, according to the sequence of user, pre-set user quantity user is chosen successively and is divided into one group,
When the quantity of remaining ungrouped user is less than the pre-set user quantity, remaining ungrouped user is divided into one group,
Obtain multiple subgroups.
5. method according to claim 1 or 2, it is characterised in that described to calculate in each virtual objects user grouping
Similarity between user, including:
Similarity in each virtual objects user grouping of parallel computation between user.
6. method according to claim 1 or 2, it is characterised in that meet in the similarity by with targeted customer pre-
If after similar users of the user of condition of similarity as the targeted customer, methods described also includes:
The preference for choosing the similar users is more than default preference threshold value, and the targeted customer did not interact
User mutual object, it is pushed to the targeted customer.
7. a kind of similar users determining device based on virtual interacting object, it is characterised in that described device includes:
User grouping acquisition module, for obtaining user grouping corresponding to each user mutual object, wherein, each user
Packet includes each user interacted and each user being present to one user mutual object with a user mutual object
Preference;
Subgroup division module, for each user grouping to be divided into multiple subgroups by default rule of classification, wherein,
For each user grouping, the quantity of user is respectively less than one user grouping in each subgroup of a user grouping
The quantity of middle user;
Virtual interacting object relating module, built for respectively each subgroup and associate virtual interacting object, obtained
Virtual objects user grouping corresponding to multiple virtual interacting objects, wherein, it is each in a subgroup for each subgroup
User to the preference of virtual interacting object corresponding to one subgroup, respectively with each user in one subgroup
It is identical to the preference of the user mutual object corresponding to user grouping corresponding to one subgroup;
Similarity calculation module, for calculating the similarity between the user in each virtual objects user grouping;
Similar users choose module, for the similarity with targeted customer to be met to the user of default condition of similarity as the mesh
Mark the similar users of user.
8. device according to claim 7, it is characterised in that described device also includes:
Preference acquisition module, for obtaining preference of each user to each user mutual object respectively;
User grouping division module, it is divided into one group for all users interacted will be present with each user mutual object, and closes
Preference corresponding to connection, obtain user grouping corresponding to each user mutual object.
9. the device according to claim 7 or 8, it is characterised in that the subgroup division module, including:
Preference sorting sub-module, for for each user grouping, user to user in the user grouping to be handed over
The preference descending arrangement of mutual object, obtains user preference degree of sequence corresponding to each user grouping;
User's sorting sub-module, will be corresponding with the preference sequence for for each preference sequence
User, it is ranked up according to the order of the preference sequence, obtains user's sequence corresponding to each user grouping;
User divides submodule, in each user's sequence, according to the sequence of user, choosing pre-set user number successively
Amount user is divided into one group, obtains multiple subgroups.
10. device according to claim 9, it is characterised in that the user divides submodule, is specifically used for:
In each user's sequence, according to the sequence of user, pre-set user quantity user is chosen successively and is divided into one group,
When the quantity of remaining ungrouped user is less than the pre-set user quantity, remaining ungrouped user is divided into one group,
Obtain multiple subgroups.
11. the device according to claim 7 or 8, it is characterised in that the similarity calculation module, be specifically used for:
Similarity in each virtual objects user grouping of parallel computation between user.
12. the device according to claim 7 or 8, it is characterised in that described device also includes:
Pushing module, the preference for choosing the similar users is more than default preference threshold value, and the target is used
The user mutual object that family did not interact, is pushed to the targeted customer.
13. a kind of electronic equipment, it is characterised in that including processor, communication interface, memory and communication bus, wherein, it is described
Processor, the communication interface, the memory complete mutual communication by the communication bus;
The memory, for depositing computer program;
The processor, during for performing the program deposited on the memory, realize any described sides of claim 1-6
Method step.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710850770.9A CN107730271B (en) | 2017-09-20 | 2017-09-20 | Similar user determination method and device based on virtual interaction object and electronic equipment |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710850770.9A CN107730271B (en) | 2017-09-20 | 2017-09-20 | Similar user determination method and device based on virtual interaction object and electronic equipment |
Publications (2)
Publication Number | Publication Date |
---|---|
CN107730271A true CN107730271A (en) | 2018-02-23 |
CN107730271B CN107730271B (en) | 2021-09-17 |
Family
ID=61206391
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201710850770.9A Active CN107730271B (en) | 2017-09-20 | 2017-09-20 | Similar user determination method and device based on virtual interaction object and electronic equipment |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN107730271B (en) |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111586001A (en) * | 2020-04-28 | 2020-08-25 | 咪咕文化科技有限公司 | Abnormal user identification method and device, electronic equipment and storage medium |
CN111797878A (en) * | 2019-04-09 | 2020-10-20 | Oppo广东移动通信有限公司 | Data processing method, data processing device, storage medium and electronic equipment |
WO2020220915A1 (en) * | 2019-04-30 | 2020-11-05 | 腾讯科技(深圳)有限公司 | Virtual object display method and apparatus, electronic device, and storage medium |
Citations (10)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102089782A (en) * | 2008-07-11 | 2011-06-08 | 摩托罗拉移动公司 | Recommender system |
CN103731738A (en) * | 2014-01-23 | 2014-04-16 | 哈尔滨理工大学 | Video recommendation method and device based on user group behavioral analysis |
CN103870547A (en) * | 2014-02-26 | 2014-06-18 | 华为技术有限公司 | Grouping processing method and device of contact persons |
CN103886001A (en) * | 2013-09-22 | 2014-06-25 | 天津思博科科技发展有限公司 | Personalized commodity recommendation system |
CN104050258A (en) * | 2014-06-15 | 2014-09-17 | 中国传媒大学 | Group recommendation method based on interest groups |
CN104935964A (en) * | 2015-06-02 | 2015-09-23 | 四川九天揽月文化传媒有限公司 | Program grouping screening and push method for intelligent television |
US20170091194A1 (en) * | 2005-03-30 | 2017-03-30 | Amazon Technologies, Inc. | Mining of user event data to identify users with common interests |
CN106874509A (en) * | 2017-03-01 | 2017-06-20 | 广州大学 | Resource recommendation method and device based on middle granularity user grouping |
CN106997347A (en) * | 2016-01-22 | 2017-08-01 | 华为技术有限公司 | Information recommendation method and server |
CN107066582A (en) * | 2017-04-14 | 2017-08-18 | 聚好看科技股份有限公司 | Realize the method and device that virtual resource is recommended |
-
2017
- 2017-09-20 CN CN201710850770.9A patent/CN107730271B/en active Active
Patent Citations (10)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20170091194A1 (en) * | 2005-03-30 | 2017-03-30 | Amazon Technologies, Inc. | Mining of user event data to identify users with common interests |
CN102089782A (en) * | 2008-07-11 | 2011-06-08 | 摩托罗拉移动公司 | Recommender system |
CN103886001A (en) * | 2013-09-22 | 2014-06-25 | 天津思博科科技发展有限公司 | Personalized commodity recommendation system |
CN103731738A (en) * | 2014-01-23 | 2014-04-16 | 哈尔滨理工大学 | Video recommendation method and device based on user group behavioral analysis |
CN103870547A (en) * | 2014-02-26 | 2014-06-18 | 华为技术有限公司 | Grouping processing method and device of contact persons |
CN104050258A (en) * | 2014-06-15 | 2014-09-17 | 中国传媒大学 | Group recommendation method based on interest groups |
CN104935964A (en) * | 2015-06-02 | 2015-09-23 | 四川九天揽月文化传媒有限公司 | Program grouping screening and push method for intelligent television |
CN106997347A (en) * | 2016-01-22 | 2017-08-01 | 华为技术有限公司 | Information recommendation method and server |
CN106874509A (en) * | 2017-03-01 | 2017-06-20 | 广州大学 | Resource recommendation method and device based on middle granularity user grouping |
CN107066582A (en) * | 2017-04-14 | 2017-08-18 | 聚好看科技股份有限公司 | Realize the method and device that virtual resource is recommended |
Non-Patent Citations (2)
Title |
---|
吴文明 等: "基于可信相似用户的服务推荐方法", 《计算机工程》 * |
田力威 等: "基于兴趣相似性的Web用户聚类方法", 《计算机与信息技术》 * |
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111797878A (en) * | 2019-04-09 | 2020-10-20 | Oppo广东移动通信有限公司 | Data processing method, data processing device, storage medium and electronic equipment |
WO2020220915A1 (en) * | 2019-04-30 | 2020-11-05 | 腾讯科技(深圳)有限公司 | Virtual object display method and apparatus, electronic device, and storage medium |
US11615570B2 (en) | 2019-04-30 | 2023-03-28 | Tencent Technology (Shenzhen) Company Limited | Virtual object display method and apparatus, electronic device, and storage medium |
CN111586001A (en) * | 2020-04-28 | 2020-08-25 | 咪咕文化科技有限公司 | Abnormal user identification method and device, electronic equipment and storage medium |
Also Published As
Publication number | Publication date |
---|---|
CN107730271B (en) | 2021-09-17 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN103823908B (en) | Content recommendation method and server based on user preference | |
CN104239571B (en) | It is a kind of to carry out using the method and apparatus recommended | |
CN104008139B (en) | The creation method and device of video index table, the recommendation method and apparatus of video | |
CN104156392B (en) | Good friend and application personalized recommendation method and system | |
CN109409928A (en) | A kind of material recommended method, device, storage medium, terminal | |
CN106294775A (en) | Content delivery method and device | |
CN105574045B (en) | Video recommendation method and server | |
CN107730271A (en) | Similar users based on virtual interacting object determine method, apparatus and electronic equipment | |
CN108510303A (en) | Advertisement placement method and device | |
CN112364202A (en) | Video recommendation method and device and electronic equipment | |
CN108363730B (en) | Content recommendation method, system and terminal equipment | |
US10762122B2 (en) | Method and device for assessing quality of multimedia resource | |
CN107491979A (en) | A kind of distribution method and device of advertisement stock | |
CN107590267A (en) | Information-pushing method and device, terminal and readable storage medium storing program for executing based on picture | |
CN106897398B (en) | Video display method and device | |
CN106503059A (en) | Displayed page method for pushing and device | |
CN108733842A (en) | Video recommendation method and device | |
CN105791902A (en) | User recommendation method and user recommendation device | |
CN105574030A (en) | Information search method and device | |
CN109558884A (en) | A kind of method, apparatus, server and medium that room classes are broadcast live | |
CN109086932A (en) | A kind of prediction technique, system and the device of media information prevalence degree | |
CN106777201B (en) | Method and device for sorting recommended data on search result page | |
CN107122805A (en) | A kind of user clustering method and apparatus | |
CN104573132B (en) | Song lookup method and device | |
CN109982158A (en) | A kind of distribution method, device and the server of UGC video |
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 |