CN106708844A - User group partitioning method and device - Google Patents
User group partitioning method and device Download PDFInfo
- Publication number
- CN106708844A CN106708844A CN201510772638.1A CN201510772638A CN106708844A CN 106708844 A CN106708844 A CN 106708844A CN 201510772638 A CN201510772638 A CN 201510772638A CN 106708844 A CN106708844 A CN 106708844A
- Authority
- CN
- China
- Prior art keywords
- label
- user
- core
- node
- level value
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/95—Retrieval from the web
- G06F16/951—Indexing; Web crawling techniques
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Databases & Information Systems (AREA)
- Data Mining & Analysis (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The embodiment of the invention provides a user group partitioning method and device. The method comprises the steps of capturing user identifiers; building a joint behavior relation between the user identifiers; obtaining a user node diagram; in the user node diagram, recognizing one or more core user groups according to the joint behavior relation; in the one or more core user groups, dividing one or more target user groups according to the joint behavior relation. On one hand, manual rule setting is avoided, different user groups have different distribution features, although individual behavior differences are large, the potential relation of the user is more stable, and the user group partitioning accuracy of the user groups is improved in a diagram-based user group partitioning mode; on the other hand, by roughly partitioning the core user groups, the data size is greatly decreased, the partitioning efficiency is improved, and the user group partitioning accuracy is improved.
Description
Technical field
The application is related to the technical field of computer disposal, more particularly to a kind of division side of user group
Method and a kind of division device of user group.
Background technology
With the high speed development of internet, online information content is sharply increased, and excessive information causes people
The part of oneself needs efficiently cannot be therefrom obtained, the service efficiency of information is reduced on the contrary.
Therefore, major websites generally by its towards user be divided into different user groups, there is provided it is more smart
The service of refinement.
In addition, in some safety detection scenes, it is also desirable to which user is divided into different user groups.
For example, in e-commerce website, lawless person increases shop by the approach such as virtual trading malice
Integration, be commonly called as " brush bore ", to keep order, website is needed the Stock discrimination of " brush is bored " out.
Now, the mode that user group divides generally has two kinds, and one kind is artificial setting rule, another
It is community discovery algorithm.
In the mode of artificial setting rule, it tends to be difficult to cover the different qualities of different groups, and,
User group it is regular various, be susceptible to change, the rule of artificial setting unavoidably can deviation, from
And cause the accuracy that user group divides relatively low.
As a example by recognizing " brush brill " colony, the conventional rule of identification " brush is bored " colony has " user's purchase
Before the similar commodity number that browses ", " user browses to the time span for placing an order ", " the multiple things of user's purchase
Interval time of product " etc..
Different " brush is bored " colony, often with different performance.As " brush an is bored " colony connects
The commodity specified directly are bought after to demand；Another " brush is bored " colony, can browse multiple similar commodity
Afterwards, then the commodity specified are bought.
Then for the two " brush bore " colonies, " the similar commodity number browsed before user's purchase " this
Performance in individual rule differs, it is difficult to be identified judging by same threshold value.
In community discovery algorithm, easily introduce in the result and the incongruent data of concrete application scene,
Cause data volume excessive, divide less efficient, the accuracy that user group divides is relatively low.
As a example by recognizing " brush bore " colony, the purchase relation between user is carried out first in modeling it is abstract,
It is simply considered that two users bought a certain part commodity and think that they have relation jointly, can causes to build
Vertical figure scale is excessive, causes to divide the low of efficiency, and exist the more use of some quantity purchases
Family is mistakenly identified as the risk of " brush is bored " user.
The content of the invention
In view of the above problems, it is proposed that the embodiment of the present application overcomes above mentioned problem or extremely to provide one kind
A kind of division methods of the user group for partially solving the above problems and a kind of corresponding user group
Division device.
In order to solve the above problems, the embodiment of the present application discloses a kind of division methods of user group, bag
Include：
Crawl ID；
The joint act relation set up between the ID, obtains user node figure；
In the user node figure, according to described one or more core customers of joint act relation recognition
Colony；
In one or more of core customer colonies, according to the joint act relation divide one or
Multiple targeted user populations.
Alternatively, the joint act relation set up between the ID, obtains user node figure
The step of include：
Search the behavioral data of the ID；
Common behavioral data is recognized from the behavioral data；
Joint act relation is set up to the ID belonging to the common behavioral data.
Alternatively, the step of behavioral data of the lookup ID includes：
Be extracted in preset time period from preset database, the behavioral data of the ID.
Alternatively, the ID to belonging to the common behavioral data sets up joint act relation
The step of include：
Type according to behavioral data configures weight to the common behavioral data；
When the weight sum is more than default weight threshold, to belonging to the common behavioral data
ID sets up joint act relation.
Alternatively, it is described in the user node figure, according to the joint act relation recognition one or
The step of multiple core customer colonies, includes：
Calculate core level value of the ID in the user node figure；
When the core level value is more than default core threshold value, determine that the core level value is corresponding
ID homing core user group.
Alternatively, the step of core level value of the calculating ID in the user node figure
Suddenly include：
The global core level value of current iteration is set；
In the user node figure, it is connected by joint act relation for each ID statistics
The quantity of ID, obtains node angle value；
In the user node figure, judge the node angle value of each ID whether less than or equal to complete
Office's core level value；
If so, then removal node angle value is less than or equal to the ID of the global core level value；
The global core level value is assigned to the ID of first removal, as the user for first removing
The core level value of mark；
In the user node figure, delete the joint act being connected with the ID of first removal and close
System, return perform it is described in the user node figure, judge each ID node angle value whether
The step of less than or equal to global core level value；
If it is not, the step of then returning to the global core level value for performing the setting current iteration, until time
The user node figure is gone through to complete.
Alternatively, the step of global core level value of the setting current iteration includes：
In iteration first, it is 1 to set initial global core level value；
Or,
In non-iteration first, Jia 1 on the basis of upper one global core level value, as the current overall situation
Core level value.
Alternatively, it is described in one or more of core customer colonies, closed according to the joint act
The step of system divides one or more targeted user populations includes：
In one or more of core customer colonies, to each ID label allocation, the mark
Signing has numerical value；
The label of each ID is transferred to connected ID；
In the label received from each ID, a label is chosen as institute according to the numerical value of label
The label for possessing；
Judge in one or more of core customer colonies, whether the label that ID is possessed is sent out
Changing；
If so, then return to perform described the label of each ID to be transferred to connected ID
Step；
If it is not, the ID that will then possess same label is divided into targeted user population.
Alternatively, it is described in one or more of core customer colonies, closed according to the joint act
The step of system divides one or more targeted user populations includes：
In one or more of core customer colonies, to each ID label allocation；
The label of each ID is transferred to connected ID；
In the label received from each ID, a label is chosen as institute according to the quantity of label
The label for possessing；
Judge in one or more of core customer colonies, whether the label that ID is possessed is sent out
Changing, or, whether currently it is less than default maximum iteration；
If so, then return to perform described the label of each ID to be transferred to connected ID
Step；
If it is not, the ID that will then possess same label is divided into targeted user population.
The embodiment of the present application also discloses a kind of division device of user group, including：
ID acquisition module, for capturing ID；
User node figure builds module, for the joint act relation set up between the ID, obtains
Obtain user's node diagram；
Core customer's Stock discrimination module, in the user node figure, according to the joint act
One or more core customer colonies of relation recognition；
Targeted user population division module, in one or more of core customer colonies, according to
The joint act relation divides one or more targeted user populations.
Alternatively, the user node figure builds module and includes：
Behavioral data searches submodule, the behavioral data for searching the ID；
Joint act data recognize submodule, for recognizing common behavior number from the behavioral data
According to；
Joint act relation setting up submodule, for the ID belonging to the common behavioral data
Set up joint act relation.
Alternatively, the behavioral data is searched submodule and is included：
Time period data searching unit, for being extracted in preset time period from preset database, institute
State the behavioral data of ID.
Alternatively, the joint act relation setting up submodule includes：
Weight dispensing unit, power is configured for the type according to behavioral data to the common behavioral data
Weight；
Relation sets up unit, for when the weight sum is more than default weight threshold, to described common
ID belonging to same behavioral data sets up joint act relation.
Alternatively, core customer's Stock discrimination module includes：
Core level value calculating sub module, for calculating the ID in the user node figure
Core level value；
Core customer colony determination sub-module, for being more than default core threshold value in the core level value
When, determine the corresponding ID homing core user group of the core level value.
Alternatively, the core level value calculating sub module includes：
Global core level value setting unit, the global core level value for setting current iteration；
Node degree Data-Statistics unit, in the user node figure, for each ID statistics
The quantity of the ID being connected by joint act relation, obtains node angle value；
Quantity comparing unit, in the user node figure, judging the node degree of each ID
Whether value is less than or equal to global core level value；If so, ID removal unit is then called, if it is not,
Then return and call the global core level value setting unit, until travel through the user node figure completing；
ID removal unit, in the user node figure, removal node angle value to be less than or waits
In the ID of the global core level value；
Core level value assignment unit, the use for the global core level value to be assigned to first removal
Family identifies, used as the core level value of the ID for first removing；
Joint act relation deletes unit, in the user node figure, deleting and first removal
The quantity comparing unit is called in the connected joint act relation of ID, return.
Alternatively, the global core level value setting unit includes：
Initial setting up subelement, is 1 in iteration first, setting initial global core level value；
Or,
Increment subelement, in non-iteration first, adding on the basis of upper one global core level value
1, as current global core level value.
Alternatively, the targeted user population division module includes：
First label configures submodule, in one or more of core customer colonies, to each
ID label allocation, the label has numerical value；
First label transmits submodule, for the label of each ID to be transferred into connected user's mark
Know；
First label chooses submodule, in the label that is received from each ID, according to label
Numerical value choose a label as the label for being possessed；
First judging submodule, for judging in one or more of core customer colonies, Yong Hubiao
Know whether possessed label changes；If so, then return calling the first label transmission submodule；
If it is not, then calling first object user group to divide submodule；
First object user group divides submodule, and the ID for will possess same label is divided into
Targeted user population.
Alternatively, the targeted user population division module includes：
Second label configures submodule, in one or more of core customer colonies, to each
ID label allocation；
Second label transmits submodule, for the label of each ID to be transferred into connected user's mark
Know；
Second label chooses submodule, in the label that is received from each ID, according to label
Quantity choose a label as the label for being possessed；
Second judging submodule, for judging in one or more of core customer colonies, Yong Hubiao
Know whether possessed label changes, or, whether currently it is less than default maximum iteration；
If so, then return calling the second label transmission submodule；If it is not, then calling the second potential user group
Body divides submodule；
Second targeted user population divides submodule, and the ID for will possess same label is divided into
Targeted user population.
The embodiment of the present application includes advantages below：
The embodiment of the present application builds user node figure by the joint act relation of user, in user node figure
In mark off core customer colony roughly, finely mark off targeted user population in core customer colony,
On the one hand, it is to avoid artificial setting rule, different customer groups has different distribution characteristics, although individual
Behavior differ greatly, but the potential relation of user is then more stable, user group based on figure divides
Mode improves the division accuracy of user group, on the other hand, core customer group is marked off by rough
Body, greatly reduces data volume, and then improves division efficiency, and the division that improve user group is accurate
Degree.
Brief description of the drawings
The step of Fig. 1 is a kind of division methods embodiment of user group of the application flow chart；
Fig. 2A-Fig. 2 C are a kind of structure exemplary plots of user node figure of the application；
Fig. 3 A- Fig. 3 D are a kind of identification exemplary plots of core customer colony of the application；
Fig. 4 is a kind of identification exemplary plot of targeted user population of the application；
Fig. 5 is a kind of structured flowchart of the division device embodiment of user group of the application.
Specific embodiment
To enable above-mentioned purpose, the feature and advantage of the application more obvious understandable, below in conjunction with the accompanying drawings
The application is described in further detail with specific embodiment.
Reference picture 1, show the application a kind of user group division methods embodiment the step of flow
Figure, specifically may include steps of：
Step 101, captures ID；
ID can represent the information of the user of determination, for example, ID
(Identity, identity number), cookie, Mac (Media Access Control, media interviews control
System) address etc..
In the embodiment of the present application, server can record the data of user by web log file, and storage exists
In database.
In dividing user groups, ID can be captured from database.
Step 102, the joint act relation set up between the ID obtains user node figure；
Joint act relation, can refer between user's (being characterized with ID) there is common behavior.
In user node figure, node on behalf user (with ID characterize), line represent node it
Between relation (i.e. joint act relation), user node figure be represent user between exist co-operate it is strong
Relational network.
In one embodiment of the application, step 102 can include following sub-step：
Sub-step S11, searches the behavioral data of the ID；
What the IP address that general web log file can record user computer is, at what time, use
Which page of website is have accessed in the case of what operating system, what browser, what display,
Whether access is successful.
But be directed to for user behavior, it is necessary to be not the IP address of user computer, operating system,
The robot datas such as browser, but user has browsed what information, the expression behaviour to its favorable rating
Etc. the behavioral data that can characterize user interest hobby.
In implementing, web log file can be filtered, obtain the behavioral data of structuring, example
Such as ID, the commodity ID that user accesses, access time, user behavior (for example click on, buy,
Evaluate etc.).
For example, web log file can be：
118.112.27.164---[24/Oct/2012:11:00:00+0800]"GET
/b.jpgcD17Mn0mdT17L2NoaW5hLmFsaWJhYmEuY29tL30mbT17R0VUfSZz
PXsyMDB9JnI9e2h0dHA6Ly9mdy50bWFsbC5jb20vP3NwbT0zLjE2OTQwNi4
xOTg0MDEufSZhPXtzaWQ9MTdjMDM2MjEtZTk2MC00NDg0LWIwNTYtZ
DJkMDcwM2NkYmE4fHN0aW1lPTEzNTEwNDc3MDU3OTZ8c2RhdGU9Mj
R8YWxpX2FwYWNoZV9pZD0xMTguMTEyLjI3LjE2NC43MjU3MzI0NzU5O
DMzMS43fGNuYT0tfSZiPXstfSZjPXtjX3NpZ25lZD0wfQ==＆pageid=7 f00000
17f00000113511803054674156071647816＆sys=ie6.0 | windowsXP | 1366*768 | zh-
Cn＆ver=43＆t=1351047705828HTTP/1.0 " 200- " Mozilla/4.0 (compatible；
MSIE 6.0；Windows NT 5.1；SV1；.NET CLR 2.0.50727)"
118.112.27.164.135104760038.6
1^sid%3D17c03621-e960-4484-b056-d2d0703cdba8%7Cstime%3D1 351047705
796%7Csdate%3D24 | cna=-^-^aid=118.112.27.164.72573247598331.7
The behavioral data of the structuring of acquisition can be after filtering：
1,b2b-1633112210,1215596848,1,07/Aug/2013:08:27:22
It should be noted that the behavior of user have it is ageing, such as summer purchase ice lolly, winter purchase plumage
Suede etc., the then foundation of joint act relation typically considers time dimension.
Therefore, in the embodiment of the present application, can be extracted in preset time period from preset database,
The behavioral data of ID.
Sub-step S12, recognizes common behavioral data from the behavioral data；
In actual applications, common behavioral data refers to identical between user's (being characterized with ID)
Behavioral data.
In e-commerce website, purchase, collection, favorable comment, the addition shopping in a period of time can be taken
The behavioral datas such as car, count in a certain time interval, commodity, common receipts that two users buy jointly
The operation note number of the commodity of Tibetan, the commodity of common favorable comment, the commodity of common addition shopping cart etc..
For example, taking the purchaser record in month, time interval is taken for one week, if buyer A is on Monday
When certain shop occur buying behavior, buyer B after three days when the shop occur buying behavior, then
There is a common behavioral data in buyer A and buyer B.
It should be noted that according to the demand of different business scene, varigrained common row can be used
It is data.
With common purchase data instance, according to application scenarios, the difference of investigation object, " common "
Relation can have flexible realization.
If " brush is bored " colony of identification particular commodity, because being single product, " common behavioral data "
Should be defined as " two users have purchased same part commodity jointly ".
If the colony for processing shop divides scene, the relation between shop is now considered, so " common
Behavioral data " may be defined as " two users have purchased arbitrary commodity in same shop ".
Sub-step S13, joint act relation is set up to the ID belonging to the common behavioral data.
In implementing, the user view intensity of different Behavior Expressions is different, for example, user buys
Commodity are intended to most by force, and collection is taken second place, and browses weaker, therefore, it can the type according to behavioral data to institute
State common behavioral data configuration weight.
In addition, pre-setting weight threshold, the setting of weight threshold is strong with the user view that the behavior expresses
It is weak to be directly proportional, typically take between 0-1.
When weight sum is more than default weight threshold, to the ID belonging to common behavioral data
Set up joint act relation.
As shown in Figure 2 A, in user node figure is built, if there is common row in user A and user B
It is data, then can be connected dotted line with user B to user A.
As shown in Figure 2 B, however, it is determined that user A and user B has stronger joint act relation, then may be used
A solid line is connected with user B with to user A.
As shown in Figure 2 C, identical operation is carried out to each user, then can builds user node figure,
As user A- user Q builds user node figure.
Step 103, in the user node figure, according to the joint act relation recognition one or many
Individual core customer colony；
Core customer colony, can refer to server towards the colony that is constituted of main users, such as behavior
The colony that more active, the more close user of association is constituted.
In implementing, the filtering of fringe node can be carried out by nomography Kcore, find out user
The node (i.e. ID) of opposite core position and their association is found in node diagram.
In one embodiment of the application, step 103 can include following sub-step：
Sub-step S21, calculates core level value of the ID in the user node figure；
In the embodiment of the present application, core level value can represent the significance level of user, core level value
It is higher, then it represents that the user is more important.
In one embodiment of the application, sub-step S21 can further include following sub-step：
Sub-step S211, sets global core level value；
In implementing, in iteration first, it is 1 that can set initial global core level value,
Assuming that setting k as 1 is k, then k=1 when initial.
In non-iteration first, can Jia 1 on the basis of upper one global core level value, as current
Global core level value, i.e. k=k+1, second iteration k=2, third time iteration k=3, by that analogy.
Sub-step S212, in the user node figure, for each ID statistics by common
The quantity of the connected ID of behavior relation, obtains node angle value；
In user node figure, certain node (i.e. ID) has N bars side, and (i.e. joint act is closed
System) connected node (i.e. ID), then its node angle value is N, and N is positive integer.
For example, as shown in Figure 2 C, node A connecting nodes B, C, D, E, F, J, then node A
Node angle value be 6；And J connecting node A of node, then the node angle value of node J is 1.
Sub-step S213, in the user node figure, judging the node angle value of each ID is
It is no less than or equal to global core level value；If so, sub-step S214 is then performed, if it is not, then returning
Sub-step S211, until travel through the user node figure completing；
Sub-step S214, in the user node figure, removal node angle value is complete less than or equal to described
The ID of office's core level value；
Sub-step S215, the global core level value is assigned to the ID of first removal, is made
It is the core level value (coreness) of ID for first removing；
Sub-step S216, in the user node figure, deletes and is connected with the ID of first removal
Joint act relation, return and perform sub-step S213, until travel through the user node figure completing.
In the embodiment of the present application, nomography Kcore supports distributed system, can process the number of magnanimity
According to.
In each iteration, node and side can be all removed, new user node figure is formed, is changed next time
Dai Zhong, i.e., processed in new user node figure.
User node figure as that shown in fig. 2 c, in iteration first, k=1, the node angle value of each node is such as
Under：
| Node angle value | Node |
| 1 | J、K、L、M、N、O、P、Q |
| 2 | E、F |
| 4 | B、C、G、H、I |
| 5 | D |
| 7 | A |
In iteration first, the node angle value of node J, K, L, M, N, O, P, Q is equal to k (1),
Therefore, node J, K, L, M, N, O, P, Q and its connected side, and assignment k are removed to section
Point J, K, L, M, N, O, P, Q, then its core level value (coreness) is 1.
User node figure as shown in Figure 3A, removal node J, K, L, M, N, O, P, Q and
After its connected side, the node angle value of each node changes, and the node angle value of such as node I is changed into 1,
The node angle value of each node is as follows：
| Node angle value | Node |
| 1 | I |
| 2 | E、F、G、H |
| 4 | B、C |
| 5 | D |
| 6 | A |
The node angle value of node I is equal to k (1), therefore, node I and its connected side are removed, and assign
Value k gives node I, then its core level value (coreness) is 1.
User node figure as shown in Figure 3 B, after removal node I and its connected side, each node
Node angle value no longer changes, also, the node angle value of all nodes is all higher than or equal to current iteration
Global core level value k (1), the node angle value of each node is as follows：
| Node angle value | Node |
| 2 | E、F、G、H |
| 4 | B、C |
| 5 | D |
| 6 | A |
Therefore, into the second wheel iteration, k=k+1=2.
In second iteration, the node angle value of node I, E, F, G, H is less than or equal to k (2),
Therefore, remove node I, E, F, G, H and its connected side, and assignment k to node I, E, F,
G, H, then its core level value (coreness) is 2.
User node figure as shown in Figure 3 C, removal node I, E, F, G, H and its connected side it
Afterwards, the node angle value of each node no longer changes, also, all nodes node angle value be all higher than or
Equal to the global core level value k (2) of current iteration, the node angle value of each node is as follows：
| Node angle value | Node |
| 3 | A、B、C、D |
Therefore, into third round iteration, k=k+1=3.
In third time iteration, the node angle value of node A, B, C, D is equal to k (3), therefore, go
Except node A, B, C, D and its connected side, and assignment k gives node A, B, C, D, then its
Core level value (coreness) is 3, and now, traverse user node diagram is completed.
User node figure as shown in Figure 3 D, the core of node J, K, L, M, N, O, P, Q, I
Heart degree value is 1 (coreness=1), and in outermost layer, the core level value of node E, F, G, H is
2 (coreness=2), in secondary outer layer, the core level value of node A, B, C, D is 3 (coreness=3),
In central core.
Sub-step S22, when the core level value is more than default core threshold value, determines the core
The corresponding ID homing core user group of degree value.
In the embodiment of the present application, core level value (coreness) can be taken more than certain core threshold value
Node set, corresponding user group for the user node figure core customer colony.
Core threshold value sets big small-scale relevant with user node figure, such as the user node of millions
Figure, the scope of core threshold value is more than 100.
In general, core customer colony is connected regardless of whether having, because according to nomography KCore
Processing procedure, core level value (coreness) more than certain core threshold value node collection credit union group
Be not in have isolated individual node into several subgraphs.
That is, the user group of several thick scopes can have been divided according to core customer here.
Certainly, in addition to nomography Kcore, core customer colony can also be recognized using other modes,
Core customer colony such as is recognized using angle value algorithm, angle value computational methods are relatively simple, angle value expression higher
The user has stronger cooperative relationship, etc., the embodiment of the present application pair with more other users
This is not any limitation as.
Step 104, in one or more core customer colonies, divides according to the joint act relation
One or more targeted user populations.
In the embodiment of the present application, can be in the base of the user group (i.e. core customer colony) of thick scope
On plinth, fine division is further carried out.
In one embodiment of the application, if the structure of user node figure is relatively simple, or, to
The dividing precision of family colony is less demanding, it is possible to use connected graph algorithm is on the basis of core customer colony
Divide targeted user population.
Wherein, in a non-directed graph, if from vertex viTo vertex vjThere is path to be connected, then claim viWith
vjBe connection, all nodes in connected graph, two-by-two between all be connection.
Such as in the scene of identification " brush is bored " colony, can be using relative during due to data modeling data cleansing
Stricter standard, now can be with the preliminary sieve series of connected graph algorithm.
In connected graph algorithm, if two users belong to different user groups, between two users not
Stronger cooperative relationship can be there are, that is, two nodes do not exist side in corresponding to user node figure.
Then in the embodiment of the present application, step 104 can include following sub-step：
Sub-step S31, in one or more of core customer colonies, configures to each ID
Label；
In implementing, for convenience of calculating, the label can be its ID, it is of course also possible to
Using other modes label allocation, such as random arrangement, as long as keeping the uniqueness of label, the application
Embodiment is not any limitation as to this.
In the embodiment of the present application, label has numerical value, such as 1,2.
Sub-step S32, connected ID is transferred to by the label of each ID；
In the embodiment of the present application, the label of each ID can be transferred to its neighbour, similarly,
The ID can receive the label of its neighbour transmission.
For example, core customer colony as shown in Figure 4, its label is transferred to node S, section by node R
Point T, receiving node S, the label of node T transmission.
Sub-step S33, in the label received from each ID, one is chosen according to the numerical value of label
Individual label is used as the label for being possessed；
In implementing, the maximum label of numerical value can be chosen, it is also possible to choose the minimum label of numerical value,
Ensure that the strategy for updating is consistent, the embodiment of the present application is not any limitation as to this.
Sub-step S34, judges in one or more of core customer colonies, ID is possessed
Label whether change；If so, then return that sub-step S32 is performed, if it is not, then performing sub-step
S35；
Sub-step S35, the ID that will possess same label is divided into targeted user population.
It is connection in same user group, between node, not because label has uniqueness
It is disconnected between node in same user group, therefore during iteration, label can be same
One user group's flowing so that the label of same user group gradually tends to stabilization, when label stabilization
When, the node with same label belongs to the corresponding user of same connected graph, i.e. node and belongs to same
User group, the label of node can be used as the identification label of the user group.
For example, it is assumed that node R, the numerical value of the label of S, T, U be respectively 1,2,
3rd, 4, select the label of numerical value minimum, then it is as follows in the process of iteration：
After the 3rd wheel iteration, the label that ID is possessed all is 1, is no longer changed, therefore,
Node R, S, T, U belong to same connected graph, and node R, the corresponding user of S, T, U belong to same
One user group.
In another embodiment of the application, if the structure of user node figure is complex, or, need
Relatively accurately to divide different user groups, it is possible to use community discovery algorithm divides different use
Family group.
Such as in the scene of identification microblogging colony, because the customer volume being related to is larger, user node figure compares
Complexity, now can obtain accuracy higher using community discovery algorithm.
In community discovery algorithm, the line belonged between the node of same user group is more dense, no
With user group node between line it is more sparse, i.e., corresponding to the node in same user group
The relation of user is more tight, can well reflect " clique " attribute of user group.
In the embodiment of the present application, community discovery algorithm supports distributed system, can process the number of magnanimity
According to.
Then in the embodiment of the present application, step 104 can include following sub-step：
Sub-step S41, in one or more of core customer colonies, configures to each ID
Label；
In implementing, for convenience of calculating, the label can be its ID, it is of course also possible to
Using other modes label allocation, such as random arrangement, as long as keeping the uniqueness of label, the application
Embodiment is not any limitation as to this.
Sub-step S42, connected ID is transferred to by the label of each ID；
Sub-step S43, in the label received from each ID, one is chosen according to the quantity of label
Individual label is used as the label for being possessed；
In implementing, the most label of quantity can be chosen, if quantity is identical, can selected at random
Take label.
Sub-step S44, judges in one or more of core customer colonies, ID is possessed
Label whether change, or, currently whether be less than default maximum iteration；If so, then
Return and perform sub-step S42, if it is not, then performing sub-step S45；
Sub-step S45, the ID that will possess same label is divided into targeted user population.
In iteration first, label can be randomly choosed, because the node of core is ined succession other many peripheries
Node, the probability that its label is arrived at random is larger, in follow-up iterative process, the mark of the node of core
Signing quantity can increase, and progressively reach stabilization.
When label stabilization or arrival maximum iteration, the node with same label belongs to same use
Family colony, the label of node can be used as the identification label of the user group.
For example, as shown in figure 4, be referred to as with the name of node the label of node, i.e. node R, S, T,
The label of U is respectively R, S, T, U, then it is as follows in the process of iteration：
After the 3rd wheel iteration, the label that ID is possessed all is R, is no longer changed, therefore,
Node R, the corresponding user of S, T, U belong to same user group.
Certainly, in addition to above-mentioned community discovery algorithm, other community discovery algorithms can also be used, such as
GN algorithms, Louvain algorithms etc., the embodiment of the present application is not any limitation as to this.
The embodiment of the present application builds user node figure by the joint act relation of user, in user node figure
In mark off core customer colony roughly, finely mark off targeted user population in core customer colony,
On the one hand, it is to avoid artificial setting rule, different customer groups has different distribution characteristics, although individual
Behavior differ greatly, but the potential relation of user is then more stable, user group based on figure divides
Mode improves the division accuracy of user group, on the other hand, core customer group is marked off by rough
Body, greatly reduces data volume, and then improves division efficiency, and the division that improve user group is accurate
Degree.
It should be noted that for embodiment of the method, in order to be briefly described, therefore it is all expressed as one it is
The combination of actions of row, but those skilled in the art should know, and the embodiment of the present application is not by described
Sequence of movement limitation because according to the embodiment of the present application, some steps can using other orders or
Person is carried out simultaneously.Secondly, those skilled in the art should also know, embodiment described in this description
Preferred embodiment is belonged to, necessary to involved action not necessarily the embodiment of the present application.
Reference picture 5, shows a kind of structured flowchart of the division device embodiment of user group of the application,
Such as lower module can specifically be included：
ID acquisition module 501, for capturing ID；
User node figure builds module 502, for the joint act relation set up between the ID,
Obtain user node figure；
Core customer's Stock discrimination module 503, in the user node figure, according to described common
Behavior relation recognizes one or more core customer colonies；
Targeted user population division module 504, in one or more of core customer colonies,
One or more targeted user populations are divided according to the joint act relation.
In one embodiment of the application, the user node figure structure module 502 can include as follows
Submodule：
Behavioral data searches submodule, the behavioral data for searching the ID；
Joint act data recognize submodule, for recognizing common behavior number from the behavioral data
According to；
Joint act relation setting up submodule, for the ID belonging to the common behavioral data
Set up joint act relation.
In an example of the embodiment of the present application, the behavioral data lookup submodule can include as follows
Unit：
Time period data searching unit, for being extracted in preset time period from preset database, institute
State the behavioral data of ID.
In an example of the embodiment of the present application, the joint act relation setting up submodule can include
Such as lower unit：
Weight dispensing unit, power is configured for the type according to behavioral data to the common behavioral data
Weight；
Relation sets up unit, for when the weight sum is more than default weight threshold, to described common
ID belonging to same behavioral data sets up joint act relation.
In one embodiment of the application, core customer's Stock discrimination module 503 can be included such as
Lower submodule：
Core level value calculating sub module, for calculating the ID in the user node figure
Core level value；
Core customer colony determination sub-module, for being more than default core threshold value in the core level value
When, determine the corresponding ID homing core user group of the core level value.
In one embodiment of the application, the core level value calculating sub module can include such as placing an order
Unit：
Global core level value setting unit, the global core level value for setting current iteration；
Node degree Data-Statistics unit, in the user node figure, for each ID statistics
The quantity of the ID being connected by joint act relation, obtains node angle value；
Quantity comparing unit, in the user node figure, judging the node degree of each ID
Whether value is less than or equal to global core level value；If so, ID removal unit is then called, if it is not,
Then return and call the global core level value setting unit, until travel through the user node figure completing；
ID removal unit, in the user node figure, removal node angle value to be less than or waits
In the ID of the global core level value；
Core level value assignment unit, the use for the global core level value to be assigned to first removal
Family identifies, used as the core level value of the ID for first removing；
Joint act relation deletes unit, in the user node figure, deleting and first removal
The quantity comparing unit is called in the connected joint act relation of ID, return.
In an example of the embodiment of the present application, the global core level value setting unit can include
Subelement：
Initial setting up subelement, is 1 in iteration first, setting initial global core level value；
Or,
Increment subelement, in non-iteration first, adding on the basis of upper one global core level value
1, as current global core level value.
In one embodiment of the application, the targeted user population division module 504 can be included such as
Lower submodule：
First label configures submodule, in one or more of core customer colonies, to each
ID label allocation, the label has numerical value；
First label transmits submodule, for the label of each ID to be transferred into connected user's mark
Know；
First label chooses submodule, in the label that is received from each ID, according to label
Numerical value choose a label as the label for being possessed；
First judging submodule, for judging in one or more of core customer colonies, Yong Hubiao
Know whether possessed label changes；If so, then return calling the first label transmission submodule；
If it is not, then calling first object user group to divide submodule；
First object user group divides submodule, and the ID for will possess same label is divided into
Targeted user population.
In another embodiment of the application, the targeted user population division module 504 can include
Following submodule：
Second label configures submodule, in one or more of core customer colonies, to each
ID label allocation；
Second label transmits submodule, for the label of each ID to be transferred into connected user's mark
Know；
Second label chooses submodule, in the label that is received from each ID, according to label
Quantity choose a label as the label for being possessed；
Second judging submodule, for judging in one or more of core customer colonies, Yong Hubiao
Know whether possessed label changes, or, whether currently it is less than default maximum iteration；
If so, then return calling the second label transmission submodule；If it is not, then calling the second potential user group
Body divides submodule；
Second targeted user population divides submodule, and the ID for will possess same label is divided into
Targeted user population.
For device embodiment, because it is substantially similar to embodiment of the method, so the comparing of description
Simply, the relevent part can refer to the partial explaination of embodiments of method.
Each embodiment in this specification is described by the way of progressive, and each embodiment is stressed
Be all difference with other embodiment, between each embodiment identical similar part mutually referring to
.
It should be understood by those skilled in the art that, the embodiment of the embodiment of the present application can be provided as method, dress
Put or computer program product.Therefore, the embodiment of the present application can using complete hardware embodiment, completely
The form of the embodiment in terms of software implementation or combination software and hardware.And, the embodiment of the present application
Can use can be situated between in one or more computers for wherein including computer usable program code with storage
The computer journey implemented in matter (including but not limited to magnetic disk storage, CD-ROM, optical memory etc.)
The form of sequence product.
In a typical configuration, the computer equipment includes one or more processors
(CPU), input/output interface, network interface and internal memory.Internal memory potentially includes computer-readable medium
In volatile memory, the shape such as random access memory (RAM) and/or Nonvolatile memory
Formula, such as read-only storage (ROM) or flash memory (flash RAM).Internal memory is computer-readable medium
Example.Computer-readable medium includes permanent and non-permanent, removable and non-removable media
Information Store can be realized by any method or technique.Information can be computer-readable instruction,
Data structure, the module of program or other data.The example of the storage medium of computer includes, but
Phase transition internal memory (PRAM), static RAM (SRAM), dynamic random is not limited to deposit
Access to memory (DRAM), other kinds of random access memory (RAM), read-only storage
(ROM), Electrically Erasable Read Only Memory (EEPROM), fast flash memory bank or other in
Deposit technology, read-only optical disc read-only storage (CD-ROM), digital versatile disc (DVD) or other
Optical storage, magnetic cassette tape, tape magnetic rigid disk storage other magnetic storage apparatus or it is any its
His non-transmission medium, can be used to store the information that can be accessed by a computing device.According to herein
Define, computer-readable medium does not include the computer readable media (transitory media) of non-standing,
Such as the data-signal and carrier wave of modulation.
The embodiment of the present application is with reference to the method according to the embodiment of the present application, terminal device (system) and meter
The flow chart and/or block diagram of calculation machine program product is described.It should be understood that can be by computer program instructions
Realize each flow and/or square frame and flow chart and/or the square frame in flow chart and/or block diagram
The combination of flow and/or square frame in figure.Can provide these computer program instructions to all-purpose computer,
The processor of special-purpose computer, Embedded Processor or other programmable data processing terminal equipments is producing
One machine so that by the computing device of computer or other programmable data processing terminal equipments
Instruction produce for realizing in one flow of flow chart or multiple one square frame of flow and/or block diagram or
The device of the function of being specified in multiple square frames.
These computer program instructions may be alternatively stored in can guide computer or other programmable datas to process
In the computer-readable memory that terminal device works in a specific way so that storage is in the computer-readable
Instruction in memory is produced and includes the manufacture of command device, and command device realization is in flow chart one
The function of being specified in flow or multiple one square frame of flow and/or block diagram or multiple square frames.
These computer program instructions can also be loaded into computer or other programmable data processing terminals set
It is standby upper so that execution series of operation steps is in terms of producing on computer or other programmable terminal equipments
The treatment that calculation machine is realized, so as to the instruction performed on computer or other programmable terminal equipments provides use
In realization in one flow of flow chart or multiple one square frame of flow and/or block diagram or multiple square frames
The step of function of specifying.
Although having been described for the preferred embodiment of the embodiment of the present application, those skilled in the art are once
Basic creative concept is known, then other change and modification can be made to these embodiments.So,
Appended claims are intended to be construed to include preferred embodiment and fall into the institute of the embodiment of the present application scope
Have altered and change.
Finally, in addition it is also necessary to explanation, herein, such as first and second or the like relational terms
It is used merely to make a distinction an entity or operation with another entity or operation, and not necessarily requires
Or imply between these entities or operation there is any this actual relation or order.And, art
Language " including ", "comprising" or any other variant thereof is intended to cover non-exclusive inclusion so that
Process, method, article or terminal device including a series of key elements not only include those key elements, and
Also include other key elements for being not expressly set out, or also include for this process, method, article or
The intrinsic key element of person's terminal device.In the absence of more restrictions, by sentence " including
It is individual ... " limit key element, it is not excluded that at the process including the key element, method, article or end
Also there is other identical element in end equipment.
Division methods above to a kind of user group provided herein and a kind of user group's draws
Separating device, is described in detail, used herein principle and embodiment party of the specific case to the application
Formula is set forth, and the explanation of above example is only intended to help and understands the present processes and its core
Thought；Simultaneously for those of ordinary skill in the art, according to the thought of the application, in specific implementation
Be will change in mode and range of application, in sum, it is right that this specification content should not be construed as
The limitation of the application.
Claims (18)
1. a kind of division methods of user group, it is characterised in that including：
Crawl ID；
The joint act relation set up between the ID, obtains user node figure；
In the user node figure, according to described one or more core customers of joint act relation recognition
Colony；
In one or more of core customer colonies, according to the joint act relation divide one or
Multiple targeted user populations.
2. method according to claim 1, it is characterised in that described to set up the ID
Between joint act relation, obtain user node figure the step of include：
Search the behavioral data of the ID；
Common behavioral data is recognized from the behavioral data；
Joint act relation is set up to the ID belonging to the common behavioral data.
3. method according to claim 2, it is characterised in that the lookup ID
Behavioral data the step of include：
Be extracted in preset time period from preset database, the behavioral data of the ID.
4. method according to claim 2, it is characterised in that described to the common behavior
The step of ID belonging to data sets up joint act relation includes：
Type according to behavioral data configures weight to the common behavioral data；
When the weight sum is more than default weight threshold, to belonging to the common behavioral data
ID sets up joint act relation.
5. the method according to claim 1 or 2 or 3 or 4, it is characterised in that described in institute
In stating user node figure, according to the step of one or more core customer colonies of joint act relation recognition
Suddenly include：
Calculate core level value of the ID in the user node figure；
When the core level value is more than default core threshold value, determine that the core level value is corresponding
ID homing core user group.
6. method according to claim 5, it is characterised in that the calculating ID
The step of core level value in the user node figure, includes：
The global core level value of current iteration is set；
In the user node figure, it is connected by joint act relation for each ID statistics
The quantity of ID, obtains node angle value；
In the user node figure, judge the node angle value of each ID whether less than or equal to complete
Office's core level value；
If so, then removal node angle value is less than or equal to the ID of the global core level value；
The global core level value is assigned to the ID of first removal, as the user for first removing
The core level value of mark；
In the user node figure, delete the joint act being connected with the ID of first removal and close
System, return perform it is described in the user node figure, judge each ID node angle value whether
The step of less than or equal to global core level value；
If it is not, the step of then returning to the global core level value for performing the setting current iteration, until time
The user node figure is gone through to complete.
7. method according to claim 6, it is characterised in that the setting current iteration it is complete
The step of office's core level value, includes：
In iteration first, it is 1 to set initial global core level value；
Or,
In non-iteration first, Jia 1 on the basis of upper one global core level value, as the current overall situation
Core level value.
8. the method according to claim 1 or 2 or 3 or 4 or 6 or 7, it is characterised in that
It is described in one or more of core customer colonies, according to the joint act relation divide one or
The step of multiple targeted user populations, includes：
In one or more of core customer colonies, to each ID label allocation, the mark
Signing has numerical value；
The label of each ID is transferred to connected ID；
In the label received from each ID, a label is chosen as institute according to the numerical value of label
The label for possessing；
Judge in one or more of core customer colonies, whether the label that ID is possessed is sent out
Changing；
If so, then return to perform described the label of each ID to be transferred to connected ID
Step；
If it is not, the ID that will then possess same label is divided into targeted user population.
9. the method according to claim 1 or 2 or 3 or 4 or 6 or 7, it is characterised in that
It is described in one or more of core customer colonies, according to the joint act relation divide one or
The step of multiple targeted user populations, includes：
In one or more of core customer colonies, to each ID label allocation；
The label of each ID is transferred to connected ID；
In the label received from each ID, a label is chosen as institute according to the quantity of label
The label for possessing；
Judge in one or more of core customer colonies, whether the label that ID is possessed is sent out
Changing, or, whether currently it is less than default maximum iteration；
If so, then return to perform described the label of each ID to be transferred to connected ID
Step；
If it is not, the ID that will then possess same label is divided into targeted user population.
10. the division device of a kind of user group, it is characterised in that including：
ID acquisition module, for capturing ID；
User node figure builds module, for the joint act relation set up between the ID, obtains
Obtain user's node diagram；
Core customer's Stock discrimination module, in the user node figure, according to the joint act
One or more core customer colonies of relation recognition；
Targeted user population division module, in one or more of core customer colonies, according to
The joint act relation divides one or more targeted user populations.
11. devices according to claim 10, it is characterised in that the user node figure builds
Module includes：
Behavioral data searches submodule, the behavioral data for searching the ID；
Joint act data recognize submodule, for recognizing common behavior number from the behavioral data
According to；
Joint act relation setting up submodule, for the ID belonging to the common behavioral data
Set up joint act relation.
12. devices according to claim 11, it is characterised in that the behavioral data searches son
Module includes：
Time period data searching unit, for being extracted in preset time period from preset database, institute
State the behavioral data of ID.
13. devices according to claim 11, it is characterised in that the joint act relation is built
Vertical submodule includes：
Weight dispensing unit, power is configured for the type according to behavioral data to the common behavioral data
Weight；
Relation sets up unit, for when the weight sum is more than default weight threshold, to described common
ID belonging to same behavioral data sets up joint act relation.
14. device according to claim 10 or 11 or 12 or 13, it is characterised in that described
Core customer's Stock discrimination module includes：
Core level value calculating sub module, for calculating the ID in the user node figure
Core level value；
Core customer colony determination sub-module, for being more than default core threshold value in the core level value
When, determine the corresponding ID homing core user group of the core level value.
15. devices according to claim 14, it is characterised in that the core level value is calculated
Submodule includes：
Global core level value setting unit, the global core level value for setting current iteration；
Node degree Data-Statistics unit, in the user node figure, for each ID statistics
The quantity of the ID being connected by joint act relation, obtains node angle value；
Quantity comparing unit, in the user node figure, judging the node degree of each ID
Whether value is less than or equal to global core level value；If so, ID removal unit is then called, if it is not,
Then return and call the global core level value setting unit, until travel through the user node figure completing；
ID removal unit, in the user node figure, removal node angle value to be less than or waits
In the ID of the global core level value；
Core level value assignment unit, the use for the global core level value to be assigned to first removal
Family identifies, used as the core level value of the ID for first removing；
Joint act relation deletes unit, in the user node figure, deleting and first removal
The quantity comparing unit is called in the connected joint act relation of ID, return.
16. devices according to claim 15, it is characterised in that the global core level value
Setting unit includes：
Initial setting up subelement, is 1 in iteration first, setting initial global core level value；
Or,
Increment subelement, in non-iteration first, adding on the basis of upper one global core level value
1, as current global core level value.
17. device according to claim 10 or 11 or 12 or 13 or 15 or 16, its feature
It is that the targeted user population division module includes：
First label configures submodule, in one or more of core customer colonies, to each
ID label allocation, the label has numerical value；
First label transmits submodule, for the label of each ID to be transferred into connected user's mark
Know；
First label chooses submodule, in the label that is received from each ID, according to label
Numerical value choose a label as the label for being possessed；
First judging submodule, for judging in one or more of core customer colonies, Yong Hubiao
Know whether possessed label changes；If so, then return calling the first label transmission submodule；
If it is not, then calling first object user group to divide submodule；
First object user group divides submodule, and the ID for will possess same label is divided into
Targeted user population.
18. device according to claim 10 or 11 or 12 or 13 or 15 or 16, its feature
It is that the targeted user population division module includes：
Second label configures submodule, in one or more of core customer colonies, to each
ID label allocation；
Second label transmits submodule, for the label of each ID to be transferred into connected user's mark
Know；
Second label chooses submodule, in the label that is received from each ID, according to label
Quantity choose a label as the label for being possessed；
Second judging submodule, for judging in one or more of core customer colonies, Yong Hubiao
Know whether possessed label changes, or, whether currently it is less than default maximum iteration；
If so, then return calling the second label transmission submodule；If it is not, then calling the second potential user group
Body divides submodule；
Second targeted user population divides submodule, and the ID for will possess same label is divided into
Targeted user population.
Priority Applications (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510772638.1A CN106708844A (en) | 2015-11-12 | 2015-11-12 | User group partitioning method and device |
| PCT/CN2016/104490 WO2017080398A1 (en) | 2015-11-12 | 2016-11-04 | Method and apparatus for dividing user group |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510772638.1A CN106708844A (en) | 2015-11-12 | 2015-11-12 | User group partitioning method and device |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN106708844A true CN106708844A (en) | 2017-05-24 |
Family
ID=58694517
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510772638.1A Pending CN106708844A (en) | 2015-11-12 | 2015-11-12 | User group partitioning method and device |
Country Status (2)
| Country | Link |
|---|---|
| CN (1) | CN106708844A (en) |
| WO (1) | WO2017080398A1 (en) |
Cited By (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107871277A (en) * | 2017-07-25 | 2018-04-03 | 平安普惠企业管理有限公司 | The method and computer-readable recording medium that server, customer relationship are excavated |
| CN108171267A (en) * | 2017-12-28 | 2018-06-15 | 广州优视网络科技有限公司 | User group partitioning method and device, information push method and device |
| CN108416645A (en) * | 2018-01-19 | 2018-08-17 | 北京三快在线科技有限公司 | A kind of recommendation method, apparatus, storage medium and equipment for user |
| CN108763359A (en) * | 2018-05-16 | 2018-11-06 | 武汉斗鱼网络科技有限公司 | A kind of usage mining method, apparatus and electronic equipment with incidence relation |
| CN108810089A (en) * | 2018-05-04 | 2018-11-13 | 微梦创科网络科技（中国）有限公司 | A kind of information-pushing method, device and storage medium |
| CN109344326A (en) * | 2018-09-11 | 2019-02-15 | 阿里巴巴集团控股有限公司 | A kind of method for digging and device of social circle |
| CN109947865A (en) * | 2018-09-05 | 2019-06-28 | 中国银联股份有限公司 | Merchant Category method and Merchant Category system |
| CN110807129A (en) * | 2019-09-29 | 2020-02-18 | 北京淇瑀信息科技有限公司 | Method and device for generating multilayer user relationship graph set and electronic equipment |
| CN111127064A (en) * | 2018-11-01 | 2020-05-08 | 百度在线网络技术（北京）有限公司 | Method and device for determining social attributes of users and electronic equipment |
| CN111309815A (en) * | 2018-12-12 | 2020-06-19 | 北京嘀嘀无限科技发展有限公司 | Method and device for processing relation map and electronic equipment |
| CN111309784A (en) * | 2020-02-13 | 2020-06-19 | 支付宝(杭州)信息技术有限公司 | Group discovery method and device |
| CN111814064A (en) * | 2020-06-24 | 2020-10-23 | 平安科技（深圳）有限公司 | Abnormal user processing method and device based on Neo4j, computer equipment and medium |
| CN111966951A (en) * | 2020-07-06 | 2020-11-20 | 东南数字经济发展研究院 | User group hierarchy dividing method based on social e-commerce transaction data |
| CN113157767A (en) * | 2021-03-24 | 2021-07-23 | 支付宝(杭州)信息技术有限公司 | Risk data monitoring method, device and equipment |
Families Citing this family (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107294974B (en) * | 2017-06-26 | 2020-10-02 | 阿里巴巴集团控股有限公司 | Method and device for identifying target group |
| CN110046805A (en) * | 2019-03-29 | 2019-07-23 | 阿里巴巴集团控股有限公司 | Risk subscribers find method and device |
| CN111831894A (en) * | 2019-04-23 | 2020-10-27 | 北京嘀嘀无限科技发展有限公司 | Information matching method and device |
| CN111651741B (en) * | 2020-06-05 | 2024-05-07 | 腾讯科技（深圳）有限公司 | User identity recognition method, device, computer equipment and storage medium |
| CN112016979B (en) * | 2020-09-08 | 2023-07-18 | 平安科技（深圳）有限公司 | User grouping method, apparatus, device and computer readable storage medium |
| CN112925800A (en) * | 2021-02-26 | 2021-06-08 | 平安普惠企业管理有限公司 | Data dependency judgment method and device, computer equipment and storage medium |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20090216581A1 (en) * | 2008-02-25 | 2009-08-27 | Carrier Scott R | System and method for managing community assets |
| CN102194149A (en) * | 2010-03-01 | 2011-09-21 | 中国人民解放军国防科学技术大学 | Community discovery method |
| CN103049496A (en) * | 2012-12-07 | 2013-04-17 | 北京百度网讯科技有限公司 | Method, apparatus and device for dividing multiple users into user groups |
| CN103678669A (en) * | 2013-12-25 | 2014-03-26 | 福州大学 | Evaluating system and method for community influence in social network |
| CN103914493A (en) * | 2013-01-09 | 2014-07-09 | 北大方正集团有限公司 | Method and system for discovering and analyzing microblog user group structure |
-
2015
- 2015-11-12 CN CN201510772638.1A patent/CN106708844A/en active Pending
-
2016
- 2016-11-04 WO PCT/CN2016/104490 patent/WO2017080398A1/en active Application Filing
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20090216581A1 (en) * | 2008-02-25 | 2009-08-27 | Carrier Scott R | System and method for managing community assets |
| CN102194149A (en) * | 2010-03-01 | 2011-09-21 | 中国人民解放军国防科学技术大学 | Community discovery method |
| CN103049496A (en) * | 2012-12-07 | 2013-04-17 | 北京百度网讯科技有限公司 | Method, apparatus and device for dividing multiple users into user groups |
| CN103914493A (en) * | 2013-01-09 | 2014-07-09 | 北大方正集团有限公司 | Method and system for discovering and analyzing microblog user group structure |
| CN103678669A (en) * | 2013-12-25 | 2014-03-26 | 福州大学 | Evaluating system and method for community influence in social network |
Cited By (21)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107871277A (en) * | 2017-07-25 | 2018-04-03 | 平安普惠企业管理有限公司 | The method and computer-readable recording medium that server, customer relationship are excavated |
| CN108171267A (en) * | 2017-12-28 | 2018-06-15 | 广州优视网络科技有限公司 | User group partitioning method and device, information push method and device |
| CN108171267B (en) * | 2017-12-28 | 2022-03-22 | 阿里巴巴（中国）有限公司 | User group division method and device and message pushing method and device |
| CN108416645A (en) * | 2018-01-19 | 2018-08-17 | 北京三快在线科技有限公司 | A kind of recommendation method, apparatus, storage medium and equipment for user |
| CN108810089B (en) * | 2018-05-04 | 2021-07-30 | 微梦创科网络科技（中国）有限公司 | Information pushing method and device and storage medium |
| CN108810089A (en) * | 2018-05-04 | 2018-11-13 | 微梦创科网络科技（中国）有限公司 | A kind of information-pushing method, device and storage medium |
| CN108763359A (en) * | 2018-05-16 | 2018-11-06 | 武汉斗鱼网络科技有限公司 | A kind of usage mining method, apparatus and electronic equipment with incidence relation |
| CN109947865A (en) * | 2018-09-05 | 2019-06-28 | 中国银联股份有限公司 | Merchant Category method and Merchant Category system |
| CN109344326B (en) * | 2018-09-11 | 2021-09-24 | 创新先进技术有限公司 | Social circle mining method and device |
| CN109344326A (en) * | 2018-09-11 | 2019-02-15 | 阿里巴巴集团控股有限公司 | A kind of method for digging and device of social circle |
| CN111127064A (en) * | 2018-11-01 | 2020-05-08 | 百度在线网络技术（北京）有限公司 | Method and device for determining social attributes of users and electronic equipment |
| CN111127064B (en) * | 2018-11-01 | 2023-08-25 | 百度在线网络技术（北京）有限公司 | Method and device for determining social attribute of user and electronic equipment |
| CN111309815A (en) * | 2018-12-12 | 2020-06-19 | 北京嘀嘀无限科技发展有限公司 | Method and device for processing relation map and electronic equipment |
| CN110807129A (en) * | 2019-09-29 | 2020-02-18 | 北京淇瑀信息科技有限公司 | Method and device for generating multilayer user relationship graph set and electronic equipment |
| CN110807129B (en) * | 2019-09-29 | 2024-02-20 | 北京淇瑀信息科技有限公司 | Method and device for generating multi-layer user relation graph set and electronic equipment |
| CN111309784A (en) * | 2020-02-13 | 2020-06-19 | 支付宝(杭州)信息技术有限公司 | Group discovery method and device |
| CN111309784B (en) * | 2020-02-13 | 2023-07-14 | 支付宝(杭州)信息技术有限公司 | Group discovery method and device |
| CN111814064A (en) * | 2020-06-24 | 2020-10-23 | 平安科技（深圳）有限公司 | Abnormal user processing method and device based on Neo4j, computer equipment and medium |
| CN111966951A (en) * | 2020-07-06 | 2020-11-20 | 东南数字经济发展研究院 | User group hierarchy dividing method based on social e-commerce transaction data |
| CN113157767A (en) * | 2021-03-24 | 2021-07-23 | 支付宝(杭州)信息技术有限公司 | Risk data monitoring method, device and equipment |
| CN113157767B (en) * | 2021-03-24 | 2022-06-07 | 支付宝(杭州)信息技术有限公司 | Risk data monitoring method, device and equipment |
Also Published As
| Publication number | Publication date |
|---|---|
| WO2017080398A1 (en) | 2017-05-18 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106708844A (en) | User group partitioning method and device | |
| CN108287864B (en) | Interest group dividing method, device, medium and computing equipment | |
| US9959563B1 (en) | Recommendation generation for infrequently accessed items | |
| CN109816482B (en) | Knowledge graph construction method, device and equipment of e-commerce platform and storage medium | |
| US20130238397A1 (en) | Data Processing Based on Online Transaction Platform | |
| US20160217479A1 (en) | Method and system for automatically recommending business prospects | |
| CN110335123B (en) | Commodity recommendation method, system, computer readable medium and device based on social e-commerce platform | |
| CN105023178B (en) | A kind of electronic commerce recommending method based on ontology | |
| CN110555172B (en) | User relationship mining method and device, electronic equipment and storage medium | |
| CN104778237A (en) | Individual recommending method and system based on key users | |
| CN108805598A (en) | Similarity information determines method, server and computer readable storage medium | |
| CN110046301B (en) | Object recommendation method and device | |
| CN111611499B (en) | Collaborative filtering method, collaborative filtering device and collaborative filtering system | |
| WO2017148273A1 (en) | Application program classification method and apparatus | |
| CN111797319B (en) | Recommendation method, recommendation device, recommendation equipment and storage medium | |
| CN111369306B (en) | Product recommendation method and device | |
| CN105894310A (en) | Personalized recommendation method | |
| CN111339435B (en) | Matrix decomposition completion hybrid recommendation method based on potential factors | |
| CN115699058A (en) | Feature interaction through edge search | |
| Al-Dhelaan et al. | Graph summarization for hashtag recommendation | |
| Meena et al. | A survey on community detection algorithm and its applications | |
| CN109299368B (en) | Method and system for intelligent and personalized recommendation of environmental information resources AI | |
| TW201828193A (en) | Method and device for dividing user group in which core user groups are first identified according to a user node diagram and then target user groups are defined among the core user groups | |
| CN110427558B (en) | Resource processing event pushing method and device | |
| WO2017036333A1 (en) | Interaction event-based webpage item recommendation method and device |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20170524 |