CN107786943A - A kind of tenant group method and computing device - Google Patents
A kind of tenant group method and computing device Download PDFInfo
- Publication number
- CN107786943A CN107786943A CN201711132674.7A CN201711132674A CN107786943A CN 107786943 A CN107786943 A CN 107786943A CN 201711132674 A CN201711132674 A CN 201711132674A CN 107786943 A CN107786943 A CN 107786943A
- Authority
- CN
- China
- Prior art keywords
- user
- label
- similarity
- users
- matrix
- 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
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W4/00—Services specially adapted for wireless communication networks; Facilities therefor
- H04W4/02—Services making use of location information
- H04W4/023—Services making use of location information using mutual or relative location information between multiple location based services [LBS] targets or of distance thresholds
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L12/00—Data switching networks
- H04L12/02—Details
- H04L12/16—Arrangements for providing special services to substations
- H04L12/18—Arrangements for providing special services to substations for broadcast or conference, e.g. multicast
- H04L12/185—Arrangements for providing special services to substations for broadcast or conference, e.g. multicast with management of multicast group membership
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L12/00—Data switching networks
- H04L12/02—Details
- H04L12/16—Arrangements for providing special services to substations
- H04L12/18—Arrangements for providing special services to substations for broadcast or conference, e.g. multicast
- H04L12/1859—Arrangements for providing special services to substations for broadcast or conference, e.g. multicast adapted to provide push services, e.g. data channels
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L67/00—Network arrangements or protocols for supporting network services or applications
- H04L67/50—Network services
- H04L67/52—Network services specially adapted for the location of the user terminal
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L67/00—Network arrangements or protocols for supporting network services or applications
- H04L67/50—Network services
- H04L67/55—Push-based network services
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W4/00—Services specially adapted for wireless communication networks; Facilities therefor
- H04W4/06—Selective distribution of broadcast services, e.g. multimedia broadcast multicast service [MBMS]; Services to user groups; One-way selective calling services
- H04W4/08—User group management
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W4/00—Services specially adapted for wireless communication networks; Facilities therefor
- H04W4/12—Messaging; Mailboxes; Announcements
Abstract
The invention discloses a kind of tenant group method, performed in computing device, computing device is connected with data storage device, connection number of each user to each wireless network is stored with data storage device, and each user is to the preference weight of each label.This method includes：The space similarity of two two users is determined to the connection number of each wireless network according to each user；The attributes similarity of two two users is determined to the preference weight of each label according to each user；Group's similarity of two two users is determined according to space similarity and attributes similarity；The group characteristics of each user vector is determined according to group's similarity of two two users；User is clustered according to the group characteristics vector of each user, user is divided into multiple groups.The present invention discloses corresponding computing device in the lump.
Description
Technical field
The present invention relates to data mining technology field, more particularly to a kind of tenant group method and computing device.
Background technology
With the development of development of Mobile Internet technology, mobile terminal turns into the important medium that people obtain information, user couple
The operation of mobile terminal can show the Behavior preference of user.User group is divided according to the Behavior preference of user,
Similar information is pushed to the user in same colony, or (was cooperateed with to targeted customer's pushed information using similar users
Filter), individualized content can be pushed to user exactly.
The more behavioural characteristics based on user of existing tenant group method (such as browse record, order record, purchasing history
Deng) and hobby (such as game, music, reading etc.) user is clustered.This method has taken into consideration only user
The feature of itself, cause its grouping result often appropriate only to specific application scenarios, it is portable poor.Further, since examine
The user characteristics considered is limited, and its grouping result is sometimes not accurate enough, it is difficult to satisfactory.
The content of the invention
Therefore, the present invention provides a kind of tenant group method and computing device, to solve or at least alleviate existing above
Problem.
According to an aspect of the present invention, there is provided a kind of tenant group method, a kind of tenant group method, in computing device
Middle execution, computing device are connected with data storage device, and it is wireless to each that each user is stored with data storage device
The connection number of network, and each user include to the preference weight of each label, this method：According to each user
The space similarity of two two users is determined to the connection number of each wireless network；Each is marked according to each user
The preference weight of label determines the attributes similarity of two two users；Determine that two is dual-purpose according to space similarity and attributes similarity
Group's similarity at family；The group characteristics of each user vector is determined according to group's similarity of two two users；According to every
The group characteristics vector of one user is clustered to user, and user is divided into multiple groups.
Alternatively, in the tenant group method according to the present invention, according to each user to each wireless network
Number is connected to include the step of determining the space similarity of two two users：According to each user to each wireless network
Number is connected to determine that adjacency matrix W, adjacency matrix W are N*N square formation, N is the quantity sum of user and wireless network, will be every
One user, each wireless network are designated as a node, the element w in adjacency matrix WijRepresent node i and node j connection
Number；The spatial signature vectors of each user are determined according to adjacency matrix W；According to the spatial signature vectors of each user
To determine the space similarity of two two users.
Alternatively, in the tenant group method according to the present invention, the sky of each user is determined according to adjacency matrix W
Between characteristic vector the step of include：Laplacian Matrix L=D-W is determined according to adjacency matrix W, wherein, D is diagonal matrix, in D
Element dii=∑jwij；Laplacian Matrix L is normalized, obtain matrix L '=D-1/2LD-1/2；To matrix L ' carry out
Eigenvalues Decomposition, characteristic value is arranged according to ascending order, has removed the preceding k outside 01Spy corresponding to individual characteristic value
Sign vector forms N*k1The first provisional matrix T1, by the first provisional matrix T1In user node corresponding to row vector conduct
The spatial signature vectors of the user.
Alternatively, in the tenant group method according to the present invention, according to the spatial signature vectors of each user come really
The step of space similarity of fixed two two users, includes：Using the cosine similarity of the spatial signature vectors of two users as this two
The space similarity of individual user.
Alternatively, in the tenant group method according to the present invention, the preference according to each user to each label
Weight includes the step of determining the attributes similarity of two two users：Preference weight according to each user to each label
To determine the attribute feature vector of each user；The category of two two users is determined according to the attribute feature vector of each user
Property similarity.
Alternatively, in the tenant group method according to the present invention, according to the attribute feature vector of each user come really
The step of attributes similarity of fixed two two users, includes：Using the cosine similarity of the attribute feature vector of two users as this two
The attributes similarity of individual user.
Alternatively, in the tenant group method according to the present invention, each user is also stored with data storage device
To the access times of each application in current slot, and application-list of labels, using being listed in-list of labels
The corresponding label of each application；User determines to the preference weight of each label according to following steps：Existed according to user
Use weight of the user to each label is determined to the access times of each application in current slot；According to user couple
Each label determines preference weight of the user to each label using weight.
Alternatively, in the tenant group method according to the present invention, user can be according to the use weight of a label
Below equation determines：
fi=α * fi-1+pi
Wherein, fiRepresent use weight of the label in current slot, fi-1Represented the label in a upper period
Using weight, α is decay factor, piFor the label current slot access times, and
Wherein, nappFor the quantity of user's used application in current slot, timesjRepresent user when current
Between in section to application j access times, βjFor Boolean factor, when corresponding to the label using j in application-list of labels, βj
=1；When not corresponding to the label using j in application-list of labels, βj=0.
Alternatively, according to the present invention tenant group method in, according to use weight of the user to each label come
Determine that user includes to the step of preference weight of each label：Preferences of the user u to label t is determined according to below equation
Weight wu,t：
Wherein, fu,tThe use weight for being user u to label t, nlabelFor the quantity of label, n is the quantity of user, ntFor
Label t using weight for 0 user quantity.
Alternatively, in the tenant group method according to the present invention, determined according to space similarity and attributes similarity
The step of group's similarity of two two users, includes：By the space similarity of two users, the weighted sum knot of attributes similarity
Group similarity of the fruit as the two users.
Alternatively, in the tenant group method according to the present invention, determined according to group's similarity of two two users every
The step of the group characteristics vector of one user includes：Group's similarity matrix is determined according to group's similarity of two two users
S；Eigenvalues Decomposition is carried out to group similarity matrix S, characteristic value is arranged according to descending order, takes preceding k2Individual feature
The corresponding characteristic vector of value forms n*k2The second provisional matrix T2, by the second provisional matrix T2In each row vector make
For the group characteristics vector of the user corresponding to the row vector.
According to another aspect of the present invention, there is provided a kind of computing device, including：At least one processor；Be stored with
The memory of programmed instruction, wherein, programmed instruction is configured as being suitable to by above-mentioned at least one computing device, programmed instruction bag
Include the instruction for performing tenant group method as described above.
According to a further aspect of the invention, there is provided a kind of readable storage medium storing program for executing for the instruction that has program stored therein, when the journey
When sequence instruction is read and performed by computing device so that the computing device tenant group method as described above.
Technique according to the invention scheme, two two users' is determined to the connection number of each wireless network according to each user
Space similarity, it is similar in terms of geographical position, social relationships, hobby that space similarity can represent two users
Degree；The attributes similarity of two two users is determined to the preference weight of each label according to each user, attributes similarity can represent
Go out similarity of two users in terms of hobby.Two are determined according to the space similarity of two two users and attributes similarity
Group's similarity of two users, the group characteristics of each user vector is determined according to group's similarity of two two users, it is right
The group characteristics vector of each user is clustered, and user is divided into multiple groups.
The tenant group method of the present invention has considered the feature of user itself, the geographic location feature of user and society
Meeting relationship characteristic, carries out a point group so that grouping result is more accurate, is applicable to a variety of fields with reference to multidimensional data to user
Scape.
Described above is only the general introduction of technical solution of the present invention, in order to better understand the technological means of the present invention,
And can be practiced according to the content of specification, and in order to allow above and other objects of the present invention, feature and advantage can
Become apparent, below especially exemplified by the embodiment of the present invention.
Brief description of the drawings
In order to realize above-mentioned and related purpose, some illustrative sides are described herein in conjunction with following description and accompanying drawing
Face, these aspects indicate the various modes that can put into practice principles disclosed herein, and all aspects and its equivalent aspect
It is intended to fall under in the range of theme claimed.Read following detailed description in conjunction with the accompanying drawings, the disclosure it is above-mentioned
And other purposes, feature and advantage will be apparent.Throughout the disclosure, identical reference generally refers to identical
Part or element.
Fig. 1 shows the schematic diagram of tenant group system 100 according to an embodiment of the invention；
Fig. 2 shows the schematic diagram of computing device 200 according to an embodiment of the invention；
Fig. 3 shows the flow chart of tenant group method 300 according to an embodiment of the invention；
Fig. 4 shows the schematic diagram of the topological structure of user according to an embodiment of the invention and wireless network；
Fig. 5 shows the schematic diagram of the adjacency matrix W and diagonal matrix D corresponding to the topological structure shown in Fig. 4；
Fig. 6 shows the flow chart of tenant group method 600 according to an embodiment of the invention.
Embodiment
The exemplary embodiment of the disclosure is more fully described below with reference to accompanying drawings.Although the disclosure is shown in accompanying drawing
Exemplary embodiment, it being understood, however, that may be realized in various forms the disclosure without should be by embodiments set forth here
Limited.On the contrary, these embodiments are provided to facilitate a more thoroughly understanding of the present invention, and can be by the scope of the present disclosure
Completely it is communicated to those skilled in the art.
Fig. 1 shows the schematic diagram of tenant group system 100 according to an embodiment of the invention.As shown in figure 1, user
Group's system 100 is divided to include computing device 200 and data storage device 120.It should be pointed out that the tenant group 100 shown in Fig. 1 is only
Exemplary, although wherein illustrate only a computing device and a data storage device, in specific practice situation
In, can there are the computing device and data storage device of varying number in tenant group system, the present invention is to tenant group system
In the quantity of included computing device and data storage device be not limited.
Computing device 200 is the equipment with communication and computing capability, and it can be implemented as server, work station etc.,
The personal computer of the configurations such as desktop computer, notebook is can be implemented as, in some cases, computing device 200
It is also implemented as the equipment such as mobile phone, tablet personal computer, intelligent wearable device.Data storage device 120 can be relationship type number
According to storehouse such as MySQL, ACCESS or non-relational database is such as NoSQL；Can reside at computing device
Local data base in 200, multiple geographical locations can also be arranged at such as HBase as distributed data base, in a word,
Data storage device 120 is used for data storage, and specific deployment of the present invention to data storage device 120, configuring condition do not limit
System.Computing device 200 can be connected with data storage device 120, and obtain the data stored in data storage device 120.
For example, the data that computing device 200 can be directly read in data storage device 120 (are set in data storage device 120 for calculating
During standby 200 local data base), internet can also be accessed by wired or wireless mode, and obtain by data-interface
Take the data in data storage device 120.
In the tenant group system 100 of the present invention, the company of user-wireless network is stored with data storage device 120
Connect relation, such as user-wireless network connection relation list shown in Fig. 1.The list includes each user to each
The connection number of wireless network, for example, first record in user-wireless network connection relation list in Fig. 1 represents, use
Connection number of the family 1 to wireless network a is 5.In addition, each user is also stored with data storage device 120 for each
The preference weight of individual label, label can be for example game, music, read, call a taxi etc., for representing the hobby of user
(or user is for the degree that is consistent of some label), user is bigger to the preference weight of a label, represents the user to this
Label is interested.For example, first record in user-label preference weight list in Fig. 1 represents that user 1 is to label 1
Preference weight be 4.2, the preference weight to label 2 is 0.5, the preference weight to label 3 is 10.1, in these three labels
In, user 1 is most interested in (or user 1 is most consistent with label 3) to label 3.
As indicated in Fig. 1 shown in the arrow index line of numeral 1., computing device 200 can read data storage device 120
Connection number of middle each the stored user to each wireless network, and each user is to the inclined of each label
Good weight, analysis calculating is carried out according to the data read, user is divided into multiple groups, and tenant group result is stored
To data storage device 120, in case he uses.For example, as shown in figure 1, computing device 200 is by calculating, it is believed that user 1, user
3rd, user 4 is more similar, and three is included into group 1；User 2 and user 5 are included into group 2；User 6 is included into group 3, etc..Fig. 1
In indicate the application scenarios that numeral arrow index line 2. shows tenant group, i.e. computing device 200 is from data storage
Tenant group result is read in device 120, different information is pushed for each group, for example, being pushed away to the user in group 1
Deliver letters breath 1, to user's pushed information 2 in group 2, to user's pushed information 3 ... in group 3, so as to realize the information content
Personalized push.
It should be pointed out that in the present invention, " user " refers to that mobile terminal, such as mobile phone, tablet personal computer, multimedia are set
Standby, intelligent wearable device etc., but not limited to this.Correspondingly, ID be mobile terminal unique mark, computing device 200
Performed " tenant group method " is actually that multiple mobile terminals are divided into multiple groups, is carried out follow-up in personalization
When holding push, and by content push to mobile terminal.For example, someone possesses 3 mobile terminals, each mobile terminal is one
Individual user, that is, this people corresponds to 3 users.Because behavior of this people on 3 mobile terminals is not quite similar, calculate
This 3 mobile terminals may be subdivided into same group by equipment 200, and identical information is pushed to 3 mobile terminals；May also
This 3 mobile terminals are divided into different groups, push the different information contents to this 3 mobile terminals respectively.
Fig. 2 shows the schematic diagram of computing device 200 according to an embodiment of the invention.In basic configuration 202,
Computing device 200 typically comprises system storage 206 and one or more processor 204.Memory bus 208 can be used
In the communication between processor 204 and system storage 206.
Depending on desired configuration, processor 204 can be any kind of processing, include but is not limited to：Microprocessor
(μ P), microcontroller (μ C), digital information processor (DSP) or any combination of them.Processor 204 can be included such as
The cache of one or more rank of on-chip cache 210 and second level cache 212 etc, processor core
214 and register 216.The processor core 214 of example can include arithmetic and logical unit (ALU), floating-point unit (FPU),
Digital signal processing core (DSP core) or any combination of them.The Memory Controller 218 of example can be with processor
204 are used together, or in some implementations, Memory Controller 218 can be an interior section of processor 204.
Depending on desired configuration, system storage 206 can be any type of memory, include but is not limited to：Easily
The property lost memory (RAM), nonvolatile memory (ROM, flash memory etc.) or any combination of them.System stores
Device 106 can include operating system 220, one or more apply 222 and routine data 224.It is actually more using 222
Bar programmed instruction, it is used to indicate that processor 204 performs corresponding operation.In some embodiments, can be arranged using 222
To cause that processor 204 is operated using routine data 224 on an operating system.
Computing device 200 can also include contributing to from various interface equipments (for example, output equipment 242, Peripheral Interface
244 and communication equipment 246) to basic configuration 202 via the communication of bus/interface controller 230 interface bus 240.Example
Output equipment 242 include graphics processing unit 248 and audio treatment unit 250.They can be configured as contributing to via
One or more A/V port 252 is communicated with the various external equipments of such as display or loudspeaker etc.Outside example
If interface 244 can include serial interface controller 254 and parallel interface controller 256, they can be configured as contributing to
Via one or more I/O port 258 and such as input equipment (for example, keyboard, mouse, pen, voice-input device, touch
Input equipment) or the external equipment of other peripheral hardwares (such as printer, scanner etc.) etc communicated.The communication of example is set
Standby 246 can include network controller 260, and it can be arranged to be easy to via one or more COM1 264 and one
The communication that other individual or multiple computing devices 262 pass through network communication link.
Network communication link can be an example of communication media.Communication media can be generally presented as in such as carrier wave
Or computer-readable instruction in the modulated data signal of other transmission mechanisms etc, data structure, program module, and can
With including any information delivery media." modulated data signal " can such signal, one in its data set or more
It is individual or it change can the mode of coding information in the signal carry out.As nonrestrictive example, communication media can be with
Include the wire medium of such as cable network or private line network etc, and it is such as sound, radio frequency (RF), microwave, infrared
(IR) the various wireless mediums or including other wireless mediums.Term computer-readable medium used herein can include depositing
Both storage media and communication media.
In the computing device 200 according to the present invention, include tenant group device 228, tenant group device using 222
228 include a plurality of programmed instruction, and routine data 224 can include each use by being got in data storage device 120
Connection number of the family to each wireless network, and each user is to the preference weight of each label.Tenant group fills
Putting 228 can indicate that processor 204 performs tenant group method 300, routine data 224 be analyzed and processed, by user
Multiple groups are divided into, realize tenant group.
Fig. 3 shows the flow chart of tenant group method 300 according to an embodiment of the invention.Method 300 is suitable to
Performed in computing device (such as aforementioned computing device 200).As shown in figure 3, method 300 starts from step S310.
In step S310, two two users' is determined to the connection number of each wireless network according to each user
Space similarity.
According to a kind of embodiment, the space similarity of two two users can come really according to following steps Step1~Step3
It is fixed：
Step1. adjacency matrix W, adjacent square are determined according to connection number of each user to each wireless network
The square formation that battle array W is N*N, N are the quantity sum of user and wireless network, and each user, each wireless network are designated as into one
Individual node, the element w in adjacency matrix WijRepresent node i and node j connection number.
The topological diagram of a user-wireless network can be obtained according to connection number of each user to each wireless network,
In the topological diagram, node is user or wireless network, and annexation of the side between two nodes, the weight on side is between two nodes
Connection number.Obviously, user will not be connected with user, wireless network and wireless network, therefore, connect user node
With user node while, be connected wireless network node and wireless network node while weight be 0, only connect user node
It is not 0 to be possible to the weight on the side of wireless network node.Adjacency matrix W is above-mentioned user-wireless network topology figure
Adjacency matrix, adjacency matrix are N*N square formation, and N is the quantity sum of user and wireless network, the element w in adjacency matrix Wij
Represent node i and the weight on node j company side, i.e. node i and node j connection number.
For example, Fig. 4 shows the topological diagram of a user-wireless network, the topological diagram includes 8 nodes, and (3 wireless
Network node and 5 user nodes), the connection number of two nodes corresponding to when upper numeral represents this.Fig. 5 is shown
Adjacency matrix W, W corresponding to Fig. 4 topological diagrams are symmetrical matrix, wijNode i and node j connection number are represented, for example, node
The connection number of 1 (wireless network node) and node 4 (user node) is 4, then the element w in adjacency matrix14=w41=4.
The geographic location feature, social relationships feature and the spy of user itself of multiple users are contained in adjacency matrix W
Sign.Generally, be connected to multiple users of same wireless network on geographical position very close to, each other may acquaintance, can
There can be identical attributive character.For example, the employee in big data company work can connect same wireless network, i.e. company
wifi.The geographical position of these people is very close to being each other Peer Relationships, may have identical feature, such as " with data
Come into contacts with " " code can be write " " interested in algorithm " etc..
Step S310 determines adjacency matrix W according to connection number of each user to each wireless network, further according to adjacent square
Battle array W determines the space similarity of two two users, in subsequent step S320~S350 according to space similarity and attributes similarity
To carry out a point group to user, considered user space characteristics (including geographic location feature and social relationships feature) and
The attributive character of user itself so that grouping result of the invention is more accurate, is applicable to several scenes.
Step2. the spatial signature vectors of each user are determined according to adjacency matrix W.According to a kind of embodiment, step
Step2 can further determine according to following steps Step21~Step23：
Step21. Laplacian Matrix L=D-W is determined according to adjacency matrix W, wherein, D is diagonal matrix, the element in D
dii=∑jwij.Matrix D is adjacency matrix W degree matrix, the element d in matrix DiiFor the weight sum on all sides of node i,
That is, element diiFor node i and the connection number sum of every other node.For example, as shown in figure 5, element d22For node 2 with
The connection number sum of every other node, i.e. the element sum of the second row in adjacency matrix W, simultaneously as adjacency matrix W is
It is poised for battle matrix, element d22Value also be adjacency matrix W in secondary series element sum.
Step22. Laplacian Matrix L is normalized, obtain matrix L '=D-1/2LD-1/2。
Step23. to matrix L ' Eigenvalues Decomposition is carried out, characteristic value is arranged according to ascending order, has removed 0
Outside preceding k1Characteristic vector corresponding to individual characteristic value forms N*k1The first provisional matrix T1, by the first provisional matrix T1In
User node corresponding to spatial signature vectors of the row vector as the user.
Matrix L ' each characteristic vector be N-dimensional column vector (i.e. N*1), by k1Individual characteristic vector (column vector) is carried out
Combination, can obtain N*k1The first provisional matrix T1, that is, the first provisional matrix T1Include N number of k1The row vector of dimension, often
One node both corresponds to a k1The row vector of dimension, the k corresponding to user node1The row vector of dimension is the space of the user
Characteristic vector.It should be pointed out that the present invention is to k1Value be not limited.
Step3. the space similarity of two two users is determined according to the spatial signature vectors of each user.According to one kind
Embodiment, the space similarity using the cosine similarity of the spatial signature vectors of two users as the two users.Cosine phase
It can be calculated like degree according to below equation：
Wherein,The spatial signature vectors of respectively two users.
Then, in step s 320, two two users are determined to the preference weight of each label according to each user
Attributes similarity.
According to a kind of embodiment, the attributes similarity of two two users can determine according to following steps：First according to each
Individual user determines the attribute feature vector of each user, the attribute feature vector of user to the preference weight of each label
In i-th of element representation user to the preference weight of i-th of label, it is clear that the length of the attribute feature vector of each user
It is identical.For example, user 1 is as shown in table 1 for the preference weight of each label：
Table 1
| ID | Label 1 | Label 2 | Label 3 | Label 4 |
| 1 | 4.2 | 0.5 | 10.1 | 0 |
Then the attribute feature vector of user 1 is [4.2,0.5,10.1,0].Then, according to the attributive character of each user
Vector determines the attributes similarity of two two users, for example, using the cosine similarity of the attribute feature vector of two users as
The attributes similarity of the two users.The computational methods of cosine similarity may be referred to aforementioned formula (1).
According to a kind of embodiment, user is not unalterable for the preference weight of each label, but constantly updates
, for example, renewal daily is once.In order to regularly update preference weight of the user for each label, according to a kind of embodiment, data
Each user is also stored with storage device 120 in current slot to the access times of each application (App), and
Using-list of labels, the label corresponding using each application is listed in list of labels, for example, " drop drop is called a taxi " application
Corresponding to " calling a taxi " label.User can determine to the preference weight of each label according to following steps Step1, Step2：
Step1. determine the user to each the access times of each application in current slot according to user
The use weight of label.It should be pointed out that the length of period can voluntarily be set by those skilled in the art, the present invention to this not
It is limited, for example, the length of period can be arranged to 1 day.
According to a kind of embodiment, user can determine to the use weight of a label according to below equation：
fi=α * fi-1+pi (2)
Wherein, fiRepresent use weight of the label in current slot, fi-1Represented the label in a upper period
Using weight, α is decay factor, and its span is [0,1], and α specific value can voluntarily be set by those skilled in the art
Put, the present invention is without limitation.piIt is the label in the access times of current slot, has：
Wherein, nappFor the quantity of user's used application in current slot, timesjRepresent user when current
Between in section to application j access times, βjFor Boolean factor, when corresponding to the label using j in application-list of labels, βj
=1；When not corresponding to the label using j in application-list of labels, βj=0.
For example, it is a user used application and access times in current slot below：
Table 2
| Using ID | a | b | c | d | e |
| Access times | 1 | 5 | 8 | 7 | 3 |
Using a~as shown in table 3 using the label corresponding to e：
Table 3
| Using ID | a | b | c | d | e |
| Tag ID | 1,2,4 | 1,3 | 2,5 | 2,5,7 | 4 |
1~label of label 7 is as shown in table 4 in the use weight of a upper period：
Table 4
| Tag ID | 1 | 2 | 3 | 4 | 5 | 6 | 7 |
| Use weight | 2.2 | 8.1 | 0.6 | 1.1 | 3.9 | 0.5 | 3 |
Label 1 illustrated below, label 2, the calculating process using weight of label 6：
It can be obtained by table 2, the user has used 5 applications altogether in current slot, respectively using a~apply e, napp
=5.From table 3, correspond to label 1 using a, using b, therefore, for label 1, β1、β2Value be 1, β3~β5's
It is worth for 0, label 1 is p in the access times of current sloti=1*1+1*5+0*8+0*7+0*3=6.Reference table 4, label 1 exist
The use weight of a upper period is 2.2, attenuation factor is arranged into 0.9, then the right to use of the label 1 in current slot
Weight is fi=0.9*2.2+6=7.98.
As shown in Table 3, label 2 is corresponded to using a, using c, using d, therefore, for label 2, β1、β3、β4Value
For 1, β2、β5Value be 0, label 2 is p in the access times of current sloti=1*1+0*5+1*8+1*7+0*3=16.With reference to
Table 4, use weight of the label 2 in a upper period is 8.1, attenuation factor is arranged into 0.9, then label 2 is in current time
The use weight of section is fi=0.9*8.1+16=23.29.
As shown in Table 3, label 6 is not corresponded to using a~using e, therefore, for label 6, β1~β5Value it is equal
For 0, access times p of the label 6 in current slotiAlso it is 0.Therefore, label 6 is upper one in the use weight of current slot
The Natural Attenuation using weight of individual period, i.e. label 6 is f in the use weight of current sloti=α * fi-1=0.9*
0.5=0.45.
Step2. preference weight of the user to each label is determined using weight to each label according to user.
According to a kind of embodiment, preference weight ws of the user u to label tu,tIt can be determined according to below equation：
Wherein, fu,tThe use weight for being user u to label t, nlabelFor the quantity of label, n is the quantity of user, ntFor
Label t using weight for 0 user quantity.The calculating effect of formula (4) is that user u is used frequently (i.e. using weight
It is larger) but other users are larger using the preference weight of label infrequently, this label is more suitable for the attribute of user
Feature.
It should be pointed out that although in figure 3, step S310, step S320 is sequentially performed successively, step S310 and step
Between rapid S320 and strict execution sequence is not present, dependence is also not present therebetween.Preferably, as shown in fig. 6, step
Rapid S310, S320 can be performed parallel, so as to accelerate calculating speed.
Then, in step S330, determine that the group of two two users is similar according to space similarity with attributes similarity
Degree.According to a kind of embodiment, group's similarity of two users is asked for the weighting of the space similarity, attributes similarity of two users
And result, i.e. group's similarity=λ1* space similarity+λ2* attributes similarity.It should be pointed out that the present invention is to λ1、λ2It is specific
Value is not limited, for example, can be by λ1、λ2Value be disposed as 0.5.
Then, in step S340, the group characteristics of each user are determined according to group's similarity of two two users
Vector.
According to a kind of embodiment, the group characteristics vector of each user can determine according to following steps：First, according to
Group's similarity of two two users determines group similarity matrix S, element s in group similarity matrix SijFor user i with
User j group's similarity.Then, Eigenvalues Decomposition is carried out to group similarity matrix S, by characteristic value according to descending
Order arranges, and takes preceding k2Characteristic vector corresponding to individual characteristic value forms n*k2The second provisional matrix T2, by the second provisional matrix
T2In each row vector as the user corresponding to the row vector group characteristics vector.It should be pointed out that the present invention is to k2's
Value is not limited.
Then, in step S350, user is clustered according to the group characteristics vector of each user, will be used
Family is divided into multiple groups.It should be pointed out that clustering algorithm density clustering such as can use DBSCAN, OPTICS is calculated
Method, the clustering algorithms based on division such as k-means, k-medoids can also be used, hierarchical clustering algorithm, net can also be used
Lattice clustering algorithm etc., the present invention to the quantity of the clustering algorithm employed in step S350 and the classification finally drawn not
It is limited.
Current most of tenant group method takes into consideration only the feature of user itself, and have ignored user geographical position,
The feature of social relationships etc..Be connected to multiple users of same Wi-Fi on geographical position very close to, each other it
Between may acquaintance, may have identical attributive character, therefore, user can represent user to the connection of wireless network
Feature in geographical position, social relationships, own interests hobby etc..The present invention is according to connection feelings of the user to wireless network
Preference weight (use feelings of the user to the preference weight of each label according to user to each application of condition and user to each label
Condition determines) a point group is carried out to user, considered the geographic location feature of user, social relationships features and itself
Attributive character so that grouping method of the invention, which can capture, to be frequently appeared in vicinal crowd, and grouping result is more
Accurately, several scenes are applicable to.
A9：Method described in A7 or 8, wherein, the use weight according to user to each label determines user
The step of preference weight of each label, is included：
Preference weight ws of the user u to label t is determined according to below equationu,t：
Wherein, fu,tThe use weight for being user u to label t, nlabelFor the quantity of label, n is the quantity of user, ntFor
Label t using weight for 0 user quantity.
A10：Method described in A1, wherein, it is described to determine two two users' according to space similarity and attributes similarity
The step of group's similarity, includes：
Using the space similarity of two users, attributes similarity weighted sum result as the two users group's phase
Like degree.
A11：Method described in A1 or 10, wherein, group's similarity according to two two users determines each use
The step of the group characteristics vector at family includes：
Group similarity matrix S is determined according to group's similarity of two two users；
Eigenvalues Decomposition is carried out to group similarity matrix S, characteristic value is arranged according to descending order, takes preceding k2
Characteristic vector corresponding to individual characteristic value forms n*k2The second provisional matrix T2, by the second provisional matrix T2In each row
Group characteristics vector of the vector as the user corresponding to the row vector.
Various technologies described herein can combine hardware or software, or combinations thereof is realized together.So as to the present invention
Method and apparatus, or some aspects of the process and apparatus of the present invention or part can take embedded tangible media, such as can
Program code (instructing) in mobile hard disk, USB flash disk, floppy disk, CD-ROM or other any machine readable storage mediums
Form, wherein when program is loaded into the machine of such as computer etc, and is performed by the machine, the machine becomes to put into practice
The equipment of the present invention.
In the case where program code performs on programmable computers, computing device generally comprises processor, processor
Readable storage medium (including volatibility and nonvolatile memory and/or memory element), at least one input unit, and extremely
A few output device.Wherein, memory is arranged to store program codes；Processor is arranged to according to the memory
Instruction in the described program code of middle storage, perform the tenant group method of the present invention.
By way of example and not limitation, computer-readable recording medium includes readable storage medium storing program for executing and communication media.Readable storage medium storing program for executing
Store the information such as computer-readable instruction, data structure, program module or other data.Communication media is typically such as to carry
The modulated message signal such as ripple or other transmission mechanisms embodies computer-readable instruction, data structure, program module or other
Data, and including any information transmitting medium.Any combination above is also included within the scope of computer-readable recording medium.
This place provide specification in, algorithm and show not with any certain computer, virtual system or other
Equipment is inherently related.Various general-purpose systems can also be used together with the example of the present invention.As described above, construct this kind of
Structure required by system is obvious.In addition, the present invention is not also directed to any certain programmed language.It should be understood that can
To realize the content of invention described herein using various programming languages, and the description done above to language-specific be for
Disclose the preferred forms of the present invention.
In the specification that this place provides, numerous specific details are set forth.It is to be appreciated, however, that the implementation of the present invention
Example can be put into practice in the case of these no details.In some instances, known method, knot is not been shown in detail
Structure and technology, so as not to obscure the understanding of this description.
Similarly, it will be appreciated that in order to simplify the disclosure and help to understand one or more of each inventive aspect,
Above in the description to the exemplary embodiment of the present invention, each feature of the invention is grouped together into single implementation sometimes
In example, figure or descriptions thereof.However, the method for the disclosure should be construed to reflect following intention：I.e. required guarantor
The application claims of shield are than the feature more features that is expressly recited in each claim.More precisely, as following
As claims reflect, inventive aspect is all features less than single embodiment disclosed above.Therefore, abide by
Thus the claims for following embodiment are expressly incorporated in the embodiment, wherein each claim is in itself
Separate embodiments as the present invention.
Those skilled in the art should be understood the module or unit or group of the equipment in example disclosed herein
Part can be arranged in equipment as depicted in this embodiment, or alternatively can be positioned at and the equipment in the example
In different one or more equipment.Module in aforementioned exemplary can be combined as a module or be segmented into addition multiple
Submodule.
Those skilled in the art, which are appreciated that, to be carried out adaptively to the module in the equipment in embodiment
Change and they are arranged in one or more equipment different from the embodiment.Can be the module or list in embodiment
Member or component be combined into a module or unit or component, and can be divided into addition multiple submodule or subelement or
Sub-component.In addition at least some in such feature and/or process or unit exclude each other, it can use any
Combination is disclosed to all features disclosed in this specification (including adjoint claim, summary and accompanying drawing) and so to appoint
Where all processes or unit of method or equipment are combined.Unless expressly stated otherwise, this specification (including adjoint power
Profit requires, summary and accompanying drawing) disclosed in each feature can be by providing the alternative features of identical, equivalent or similar purpose come generation
Replace.
In addition, it will be appreciated by those of skill in the art that although some embodiments described herein include other embodiments
In included some features rather than further feature, but the combination of the feature of different embodiments means in of the invention
Within the scope of and form different embodiments.For example, in the following claims, embodiment claimed is appointed
One of meaning mode can use in any combination.
In addition, be described as herein can be by the processor of computer system or by performing for some in the embodiment
The method or the combination of method element that other devices of the function are implemented.Therefore, have and be used to implement methods described or method
The processor of the necessary instruction of element forms the device for implementing this method or method element.In addition, device embodiment
Element described in this is the example of following device：The device is used to implement as in order to performed by implementing the element of the purpose of the invention
Function.
As used in this, unless specifically stated so, come using ordinal number " first ", " second ", " the 3rd " etc.
Description plain objects are merely representative of the different instances for being related to similar object, and are not intended to imply that the object being so described must
Must have the time it is upper, spatially, in terms of sequence or given order in any other manner.
Although describing the present invention according to the embodiment of limited quantity, above description, the art are benefited from
It is interior it is clear for the skilled person that in the scope of the present invention thus described, it can be envisaged that other embodiments.Additionally, it should be noted that
The language that is used in this specification primarily to readable and teaching purpose and select, rather than in order to explain or limit
Determine subject of the present invention and select.Therefore, in the case of without departing from the scope and spirit of the appended claims, for this
Many modifications and changes will be apparent from for the those of ordinary skill of technical field.For the scope of the present invention, to this
The done disclosure of invention is illustrative and be not restrictive, and it is intended that the scope of the present invention be defined by the claims appended hereto.
Claims (10)
1. a kind of tenant group method, is performed in computing device, the computing device is connected with data storage device, the number
According to being stored with connection number of each user to each wireless network in storage device, and each user is to each
The preference weight of label, methods described include：
The space similarity of two two users is determined to the connection number of each wireless network according to each user；
The attributes similarity of two two users is determined to the preference weight of each label according to each user；
Group's similarity of two two users is determined according to space similarity and attributes similarity；
The group characteristics of each user vector is determined according to group's similarity of two two users；
User is clustered according to the group characteristics vector of each user, user is divided into multiple groups.
2. the method for claim 1, wherein connection number according to each user to each wireless network
To include the step of determining the space similarity of two two users：
Determine that adjacency matrix W, adjacency matrix W are N*N's according to connection number of each user to each wireless network
Square formation, N are the quantity sum of user and wireless network, and each user, each wireless network are designated as into a node, adjacent
Element w in matrix WijRepresent node i and node j connection number；
The spatial signature vectors of each user are determined according to adjacency matrix W；
The space similarity of two two users is determined according to the spatial signature vectors of each user.
3. method as claimed in claim 2, wherein, the space characteristics that each user is determined according to adjacency matrix W
The step of vector includes：
Laplacian Matrix L=D-W is determined according to adjacency matrix W, wherein, D is diagonal matrix, the element d in Dii=∑jwij；
Laplacian Matrix L is normalized, obtain matrix L '=D-1/2LD-1/2；
To matrix L ' Eigenvalues Decomposition is carried out, characteristic value is arranged according to ascending order, has removed the preceding k outside 01It is individual
Characteristic vector corresponding to characteristic value forms N*k1The first provisional matrix T1, by the first provisional matrix T1In user node institute
Spatial signature vectors of the corresponding row vector as the user.
4. method as claimed in claim 2 or claim 3, wherein, the spatial signature vectors according to each user determine two
The step of space similarity of two users, includes：
Space similarity using the cosine similarity of the spatial signature vectors of two users as the two users.
5. the method for claim 1, wherein it is described according to each user to the preference weight of each label come really
The step of attributes similarity of fixed two two users, includes：
The attribute feature vector of each user is determined to the preference weight of each label according to each user；
The attributes similarity of two two users is determined according to the attribute feature vector of each user.
6. method as claimed in claim 5, wherein, the attribute feature vector according to each user determines that two is dual-purpose
The step of attributes similarity at family, includes：
Attributes similarity using the cosine similarity of the attribute feature vector of two users as the two users.
7. the method as described in claim 5 or 6, wherein, each user is also stored with the data storage device and is being worked as
To the access times of each application, and application-list of labels in the preceding period, listed in the application-list of labels
The corresponding label of each application；
The user determines to the preference weight of each label according to following steps：
Use of the user to each label is determined to the access times of each application in current slot according to user
Weight；
Preference weight of the user to each label is determined using weight to each label according to user.
8. method as claimed in claim 7, wherein, user can be true according to below equation to the use weight of a label
It is fixed：
fi=α * fi-1+pi
Wherein, fiRepresent use weight of the label in current slot, fi-1Represented use of the label in a upper period
Weight, α are decay factor, piFor the label current slot access times, and
<mrow>
<msub>
<mi>p</mi>
<mi>i</mi>
</msub>
<mo>=</mo>
<munderover>
<mo>&Sigma;</mo>
<mrow>
<mi>j</mi>
<mo>=</mo>
<mn>1</mn>
</mrow>
<msub>
<mi>n</mi>
<mrow>
<mi>a</mi>
<mi>p</mi>
<mi>p</mi>
</mrow>
</msub>
</munderover>
<mrow>
<mo>(</mo>
<msub>
<mi>&beta;</mi>
<mi>j</mi>
</msub>
<mo>*</mo>
<msub>
<mi>times</mi>
<mi>j</mi>
</msub>
<mo>)</mo>
</mrow>
</mrow>
Wherein, nappFor the quantity of user's used application in current slot, timesjRepresent user in current slot
The interior access times to application j, βjFor Boolean factor, when corresponding to the label using j in application-list of labels, βj=1；
When not corresponding to the label using j in application-list of labels, βj=0.
9. a kind of computing device, including：
At least one processor；With
Have program stored therein the memory of instruction, wherein, described program instruction is configured as being suitable to by least one processor
Perform, described program instruction includes being used for the instruction for performing the tenant group method as any one of claim 1-8.
10. a kind of readable storage medium storing program for executing for the instruction that has program stored therein, when described program instruction is read and performed by computing device,
So that tenant group method of the computing device as any one of claim 1-8.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711132674.7A CN107786943B (en) | 2017-11-15 | 2017-11-15 | User grouping method and computing device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711132674.7A CN107786943B (en) | 2017-11-15 | 2017-11-15 | User grouping method and computing device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN107786943A true CN107786943A (en) | 2018-03-09 |
| CN107786943B CN107786943B (en) | 2020-09-01 |
Family
ID=61433346
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201711132674.7A Active CN107786943B (en) | 2017-11-15 | 2017-11-15 | User grouping method and computing device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107786943B (en) |
Cited By (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108681581A (en) * | 2018-05-11 | 2018-10-19 | 重庆工商职业学院 | A kind of collaborative filtering method and system of employment data |
| CN109872722A (en) * | 2019-01-17 | 2019-06-11 | 珠海格力电器股份有限公司 | A kind of voice interactive method, device, storage medium and air-conditioning |
| CN110083791A (en) * | 2019-05-05 | 2019-08-02 | 北京三快在线科技有限公司 | Target group detection method, device, computer equipment and storage medium |
| CN110335070A (en) * | 2019-06-21 | 2019-10-15 | 北京淇瑀信息科技有限公司 | A kind of method, apparatus and electronic equipment of the user group extension based on WIFI |
| CN110598122A (en) * | 2018-06-13 | 2019-12-20 | 百度在线网络技术（北京）有限公司 | Social group mining method, device, equipment and storage medium |
| CN110855487A (en) * | 2019-11-13 | 2020-02-28 | 腾讯科技（深圳）有限公司 | Network user similarity management method, device and storage medium |
| CN110874607A (en) * | 2018-08-31 | 2020-03-10 | 浙江大学 | Clustering method and device for network nodes |
| CN111159576A (en) * | 2019-12-31 | 2020-05-15 | 亚信科技（中国）有限公司 | User classification method, device and system |
| CN111831890A (en) * | 2019-04-17 | 2020-10-27 | 腾讯科技（深圳）有限公司 | User similarity generation method and device, storage medium and computer equipment |
| WO2021012482A1 (en) * | 2019-07-23 | 2021-01-28 | 平安科技（深圳）有限公司 | Method and device for generating group interest tag, computer device, and storage medium |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2005182611A (en) * | 2003-12-22 | 2005-07-07 | Fuji Xerox Co Ltd | Document classification method and device |
| CN101547162A (en) * | 2008-03-28 | 2009-09-30 | 国际商业机器公司 | Method and device for tagging user based on user state information |
| CN101685458A (en) * | 2008-09-27 | 2010-03-31 | 华为技术有限公司 | Recommendation method and system based on collaborative filtering |
| CN103106279A (en) * | 2013-02-21 | 2013-05-15 | 浙江大学 | Clustering method simultaneously based on node attribute and structural relationship similarity |
| CN103377262A (en) * | 2012-04-28 | 2013-10-30 | 国际商业机器公司 | Method and device for grouping users |
| CN103593349A (en) * | 2012-08-14 | 2014-02-19 | 中国科学院沈阳自动化研究所 | Movement position analysis method in sense network environment |
| CN105718951A (en) * | 2016-01-21 | 2016-06-29 | 腾讯科技（深圳）有限公司 | User similarity estimation method and system |
| CN106056455A (en) * | 2016-06-02 | 2016-10-26 | 南京邮电大学 | Group and place recommendation method based on location and social relationship |
-
2017
- 2017-11-15 CN CN201711132674.7A patent/CN107786943B/en active Active
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2005182611A (en) * | 2003-12-22 | 2005-07-07 | Fuji Xerox Co Ltd | Document classification method and device |
| CN101547162A (en) * | 2008-03-28 | 2009-09-30 | 国际商业机器公司 | Method and device for tagging user based on user state information |
| CN101685458A (en) * | 2008-09-27 | 2010-03-31 | 华为技术有限公司 | Recommendation method and system based on collaborative filtering |
| CN103377262A (en) * | 2012-04-28 | 2013-10-30 | 国际商业机器公司 | Method and device for grouping users |
| CN103593349A (en) * | 2012-08-14 | 2014-02-19 | 中国科学院沈阳自动化研究所 | Movement position analysis method in sense network environment |
| CN103106279A (en) * | 2013-02-21 | 2013-05-15 | 浙江大学 | Clustering method simultaneously based on node attribute and structural relationship similarity |
| CN105718951A (en) * | 2016-01-21 | 2016-06-29 | 腾讯科技（深圳）有限公司 | User similarity estimation method and system |
| CN106056455A (en) * | 2016-06-02 | 2016-10-26 | 南京邮电大学 | Group and place recommendation method based on location and social relationship |
Cited By (17)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108681581A (en) * | 2018-05-11 | 2018-10-19 | 重庆工商职业学院 | A kind of collaborative filtering method and system of employment data |
| CN110598122A (en) * | 2018-06-13 | 2019-12-20 | 百度在线网络技术（北京）有限公司 | Social group mining method, device, equipment and storage medium |
| CN110598122B (en) * | 2018-06-13 | 2022-04-01 | 百度在线网络技术（北京）有限公司 | Social group mining method, device, equipment and storage medium |
| CN110874607A (en) * | 2018-08-31 | 2020-03-10 | 浙江大学 | Clustering method and device for network nodes |
| CN109872722A (en) * | 2019-01-17 | 2019-06-11 | 珠海格力电器股份有限公司 | A kind of voice interactive method, device, storage medium and air-conditioning |
| CN109872722B (en) * | 2019-01-17 | 2021-08-31 | 珠海格力电器股份有限公司 | Voice interaction method and device, storage medium and air conditioner |
| CN111831890A (en) * | 2019-04-17 | 2020-10-27 | 腾讯科技（深圳）有限公司 | User similarity generation method and device, storage medium and computer equipment |
| CN111831890B (en) * | 2019-04-17 | 2024-02-13 | 腾讯科技（深圳）有限公司 | User similarity generation method, device, storage medium and computer equipment |
| CN110083791B (en) * | 2019-05-05 | 2020-04-24 | 北京三快在线科技有限公司 | Target group detection method and device, computer equipment and storage medium |
| CN110083791A (en) * | 2019-05-05 | 2019-08-02 | 北京三快在线科技有限公司 | Target group detection method, device, computer equipment and storage medium |
| CN110335070B (en) * | 2019-06-21 | 2024-01-26 | 北京淇瑀信息科技有限公司 | WIFI-based user group extension method and device and electronic equipment |
| CN110335070A (en) * | 2019-06-21 | 2019-10-15 | 北京淇瑀信息科技有限公司 | A kind of method, apparatus and electronic equipment of the user group extension based on WIFI |
| WO2021012482A1 (en) * | 2019-07-23 | 2021-01-28 | 平安科技（深圳）有限公司 | Method and device for generating group interest tag, computer device, and storage medium |
| CN110855487B (en) * | 2019-11-13 | 2022-02-18 | 腾讯科技（深圳）有限公司 | Network user similarity management method, device and storage medium |
| CN110855487A (en) * | 2019-11-13 | 2020-02-28 | 腾讯科技（深圳）有限公司 | Network user similarity management method, device and storage medium |
| CN111159576B (en) * | 2019-12-31 | 2023-08-11 | 亚信科技（中国）有限公司 | User classification method, device and system |
| CN111159576A (en) * | 2019-12-31 | 2020-05-15 | 亚信科技（中国）有限公司 | User classification method, device and system |
Also Published As
| Publication number | Publication date |
|---|---|
| CN107786943B (en) | 2020-09-01 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107786943A (en) | A kind of tenant group method and computing device | |
| US11227190B1 (en) | Graph neural network training methods and systems | |
| Kosinski et al. | Mining big data to extract patterns and predict real-life outcomes. | |
| CN110866181B (en) | Resource recommendation method, device and storage medium | |
| US10354184B1 (en) | Joint modeling of user behavior | |
| CN107704485A (en) | A kind of position recommends method and computing device | |
| CN111179031B (en) | Training method, device and system for commodity recommendation model | |
| CN109344314B (en) | Data processing method and device and server | |
| CN104331411B (en) | The method and apparatus of recommended project | |
| CN110008397B (en) | Recommendation model training method and device | |
| CN106874503B (en) | Method and device for acquiring recommended data | |
| US20220138502A1 (en) | Graph neural network training methods and systems | |
| CN104598601B (en) | A kind of method, apparatus classified to user and content and computing device | |
| CN110503506A (en) | Item recommendation method, device and medium based on score data | |
| CN112528164B (en) | User collaborative filtering recall method and device | |
| CN106452809B (en) | Data processing method and device | |
| CN107341233A (en) | A kind of position recommends method and computing device | |
| CN110427564A (en) | Deep learning is integrated into broad sense additivity melange effect (GAME) frame | |
| CN105308591A (en) | Dynamics of tie strength from social interaction | |
| CN110390014A (en) | A kind of Topics Crawling method, apparatus and storage medium | |
| CN111047410B (en) | Recommendation method, recommendation device, terminal equipment and storage medium | |
| CN113704620A (en) | User label updating method, device, equipment and medium based on artificial intelligence | |
| CN109819002B (en) | Data pushing method and device, storage medium and electronic device | |
| CN115495663A (en) | Information recommendation method and device, electronic equipment and storage medium | |
| CN114417161A (en) | Virtual article time sequence recommendation method, device, medium and equipment based on special-purpose map |
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 | ||
| CP02 | Change in the address of a patent holder |
Address after: 100027 302, 3 / F, aviation service building, Dongzhimen street, Dongcheng District, Beijing Patentee after: BEIJING TENDCLOUD TIANXIA TECHNOLOGY Co.,Ltd. Address before: Room 2104, 2 / F, building 4, 75 Suzhou street, Haidian District, Beijing 100027 Patentee before: BEIJING TENDCLOUD TIANXIA TECHNOLOGY Co.,Ltd. |
|
| CP02 | Change in the address of a patent holder | ||
| PE01 | Entry into force of the registration of the contract for pledge of patent right |
Denomination of invention: A user clustering method and computing equipment Effective date of registration: 20220622 Granted publication date: 20200901 Pledgee: Beijing Yizhuang International Financing Guarantee Co.,Ltd. Pledgor: BEIJING TENDCLOUD TIANXIA TECHNOLOGY CO.,LTD. Registration number: Y2022980008462 |
|
| PE01 | Entry into force of the registration of the contract for pledge of patent right | ||
| PC01 | Cancellation of the registration of the contract for pledge of patent right |
Date of cancellation: 20230721 Granted publication date: 20200901 Pledgee: Beijing Yizhuang International Financing Guarantee Co.,Ltd. Pledgor: BEIJING TENDCLOUD TIANXIA TECHNOLOGY CO.,LTD. Registration number: Y2022980008462 |
|
| PC01 | Cancellation of the registration of the contract for pledge of patent right |