CN104504264A - Virtual person building method and device - Google Patents
Virtual person building method and device Download PDFInfo
- Publication number
- CN104504264A CN104504264A CN201410814330.4A CN201410814330A CN104504264A CN 104504264 A CN104504264 A CN 104504264A CN 201410814330 A CN201410814330 A CN 201410814330A CN 104504264 A CN104504264 A CN 104504264A
- Authority
- CN
- China
- Prior art keywords
- node
- account
- visual human
- value
- similarity
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Classifications
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16Z—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS, NOT OTHERWISE PROVIDED FOR
- G16Z99/00—Subject matter not provided for in other main groups of this subclass
Landscapes
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention relates to a virtual person building method and device based on behavior log. The virtual method includes extracting account number, log-in time, log-in terminal information from the behavior log, calculating the similarity among accounts according to the emergence of collaboration between accounts, constructing a connected graph according to characterization of accounts at the nodes, and clustering the connecting nodes in the graph and building a virtual person according to the result of clustering. The similarity among the accounts is characterized with the length of the edge between nodes. The shorter the edge between nodes is, the higher the similarity among the accounts represented by the nodes is. The invention also relates to a virtual person building device. The method and device builds a virtual person based on the behavior log, the complex rate is low, the accuracy is high and the method and device is suitable for processing large data.
Description
Technical field
The present invention relates to technical field of data processing, particularly relate to a kind of visual human's method for building up and device of Behavior-based control daily record.
Background technology
Current, instant messaging, Email, online game, P2P software download, network forum, E-Recruit, e-commerce transaction, the various network services such as the predetermined air ticket hotel of network bring great convenience to the life of the network user.Various network service distributes an account number generally can to each user, account is associated with the log-on message of user and in order to record each user and to identify, the instant communication number (as QQ account) of the such as network user or e-mail address, online game account number, forum logs in account number, and P2P software account number etc.
Each network user has the various account of type, the account data of the flood tide that a large amount of network users then brings, and concerning relevant departments, effective supervising the network user profile becomes difficult task.For effective supervising the network user profile, realize the analysis to network account attaching relation, namely which account number belongs to same person (visual human), has now become the problem needing solution badly.
Prior art, when the problem in the face of building visual human, is attributed to attributes match mode mostly.The scheme of attributes match is roughly as follows:
A) rule of specified network account attributes coupling, with which attribute mates in which kind of situation, and the match is successful accordingly decision method.Such as, when coupling QQ account number and Taobao's account number, if the editing distance (editdistance) of " name " of two account numbers and " contact method " two fields is all less than 3, then the match is successful to think these two account numbers.
B) according to the situation of attributes match, the degree (similarity) belonging to same person between account number is built.And finally telling which account number according to similarity belongs to same person.Such as, in upper example, then think belong to same person as long as the match is successful.
But, there is following situation in real life:
1. often there is the situation that attribute lacks in account data, such as, only fill in part property value during account registration.
2. dissimilar account data, total attribute is few.And in total attribute, not necessarily can be used for attributes match.
3. dissimilar account data, different to the attribute of same semanteme, need alignment, which in turns increases difficulty.Such as in category-A account number, the field that name is corresponding is exactly " name " this field, but in category-B account number, name is actually and represents by " surname " and " name " two fields.
4., in actual account number data, the confidence level of property value is not very high.Such as, in default of real-name authentication, the false situation of identification card number may be there is.
5. need the comparison carrying out properties level, complexity is higher.
These situations make that the process of attributes match is complicated, calculated amount large and actual result is undesirable, and time especially for mass data process, accuracy is lower.
Summary of the invention
Therefore, the object of the present invention is to provide a kind of visual human's method for building up of Behavior-based control daily record, solve the visual human brought because account number type is various etc. and build the low problem of complexity, accuracy.
Another object of the present invention is to visual human's apparatus for establishing that a kind of Behavior-based control daily record is provided, solve the visual human brought because account number type is various etc. and build the low problem of complexity, accuracy.
For achieving the above object, the invention provides a kind of visual human's method for building up, comprise the steps:
Extract account and the landing time corresponding with account in subordinate act daily record, log in end message;
The similarity between account is calculated according to situation about occurring collaborative between account, construct the connected graph characterizing account with node, and characterizing the similarity between account with the length on the limit between node, the limit between node is shorter, and between the account that node characterizes, similarity is higher;
Cluster is carried out to the node in described connected graph, sets up visual human according to cluster result.
Wherein, the factor also introduced between account beyond collaborative situation about occurring calculates the similarity between described account.
Wherein, the process that the node in described connected graph carries out cluster is comprised the steps:
Local density Rho, the Rho that obtain each node are respectively defined as the number of length lower than the adjacent side of certain predefine value Dc of this node of connection;
Obtain the dispersion Delta of each node respectively, Delta is defined as the length of side of most minor face in the adjacent side of all connections of this node higher Rho value neighbor node; If there is not such neighbor node, then get the length of side of the longest adjacent side of this node.
Be the Centroid of class higher than the node identification of predetermined threshold value R_T and D_T respectively by Rho value and Delta value;
Non-central node is classified as the shortest and Rho value of this non-central nodal distance higher than this non-central node Centroid belonging to class;
Each node of same item together forms a visual human, namely belongs to same visual human.
Wherein, K-Means method or hierarchy clustering method is adopted to carry out cluster to the node in described connected graph.
Wherein, also comprise and merge all visual humans and the account corresponding with visual human and become Virtual Human Data storehouse.
Present invention also offers a kind of visual human's apparatus for establishing, comprising:
Information extraction unit, for extracting account and the landing time corresponding with account, logging in end message in subordinate act daily record;
Connected graph tectonic element, for calculating the similarity between account according to situation about occurring collaborative between account, construct the connected graph characterizing account with node, and characterize the similarity between account with the length on the limit between node, limit between node is shorter, and between the account that node characterizes, similarity is higher;
Visual human sets up unit, for carrying out cluster to the node in described connected graph, sets up visual human according to cluster result.
Wherein, also comprise external model and introduce unit, calculate the similarity between described account for the factor introduced between account beyond collaborative situation about occurring.
Wherein, the process that the node in described connected graph carries out cluster is comprised the steps:
Local density Rho, the Rho that obtain each node are respectively defined as the number of length lower than the adjacent side of certain predefine value Dc of this node of connection;
Obtain the dispersion Delta of each node respectively, Delta is defined as the length of side of most minor face in the adjacent side of all connections of this node higher Rho value neighbor node; If there is not such neighbor node, then get the length of side of the longest adjacent side of this node.
Be the Centroid of class higher than the node identification of predetermined threshold value R_T and D_T respectively by Rho value and Delta value;
Non-central node is classified as the shortest and Rho value of this non-central nodal distance higher than this non-central node Centroid belonging to class;
Each node of same item together forms a visual human.
Wherein, K-Means method or hierarchy clustering method is adopted to carry out cluster to the node in described connected graph.
Wherein, also comprise visual human's merge cells, for merging all visual humans and the account corresponding with visual human becomes Virtual Human Data storehouse.
In sum, visual human is set up in visual human's method for building up of the present invention and the daily record of device Behavior-based control, and complexity is low, and accuracy rate is high, is suitable for processing large data.
Accompanying drawing explanation
In accompanying drawing,
Fig. 1 is the process flow diagram of visual human's method for building up one of the present invention preferred embodiment;
Fig. 2 is the logical schematic of visual human's method for building up one of the present invention preferred embodiment;
Fig. 3 is the Rho value-Delta Distribution value schematic diagram in visual human's method for building up one of the present invention preferred embodiment;
Fig. 4 is the structural representation of visual human's apparatus for establishing one of the present invention preferred embodiment.
Embodiment
Below in conjunction with accompanying drawing, by the specific embodiment of the present invention describe in detail, will make technical scheme of the present invention and beneficial effect apparent.
See Fig. 1, it is the process flow diagram of visual human's method for building up one of the present invention preferred embodiment.Key step of the present invention comprises:
Extract account and the landing time corresponding with account in subordinate act daily record, log in end message;
The similarity between account is calculated according to situation about occurring collaborative between account, construct the connected graph characterizing account with node, and characterizing the similarity between account with the length on the limit between node, the limit between node is shorter, and between the account that node characterizes, similarity is higher;
Cluster is carried out to the node in described connected graph, sets up visual human according to cluster result.
The present invention can also comprise and merge the step that all visual humans and the account corresponding with visual human become Virtual Human Data storehouse.
The visual human brought because account number type is various etc. for reply builds the practical problemss such as complexity, accuracy be low, the present invention proposes a kind of analytical approach of Behavior-based control daily record.User behaviors log have recorded the situation of network user's application network service, can gather from server end, user terminal etc.The method is based on the following observation to reality:
1., in a period of time, same station terminal there is movable account number may belong to same person.We claim multiple account number within certain a period of time to have activity in same terminal, are the collaborative appearance of these account numbers.
2. many account numbers situation of working in coordination with appearance more approximate-such as number of times is more, it is larger that these account numbers belong to same person possibility (claiming, similarity).
3., in multiple account numbers that unique user has, always there is part account number to use more frequent.
4., between the part account number of different user, even if there be collaborative appearance once in a while, its collaborative situation about occurring can not be more approximate than situation about occurring collaborative between each account number of user oneself.
See Fig. 2, it is the logical schematic of visual human's method for building up one of the present invention preferred embodiment.
Final steps in this preferred embodiment comprises:
Step 1. by abstract for recording in user behaviors log be the [time, terminal, account number], thus obtain comprising timestamp, the data of account ID and Termination ID, when thus learn which account number has activity in which terminal, by adding up in this account a period of time to each account and other account numbers had movable collaborative occurrence number in same terminal, the collaborative number of times occurred between account can be drawn.
" number of times " is a kind of mode weighing " situation ", and the saying adopting " number of times " in this embodiment is only for the purpose of simplifying the description.In fact, the information such as period can also be added and weigh " situation "-such as together as weights, the weight of the collaborative appearance of quitting time can slightly overweight work hours-work hours more may shared computer terminal.
Step 2. works in coordination with the observation of the situation of appearance based on above-mentioned account, calculates the similarity between account number.If be abstracted into connected graph, then the node on behalf account number in connected graph, the length on limit characterizes the similarity between account number.Under normal circumstances, similarity is higher, and limit is shorter.
Step 3., can using same for the matching result of corresponding model as the factor affecting edge lengths if any other models, such as attributes match.
After step 4. obtains above-mentioned figure, can calculate as follows, show which account number belongs to same person:
Step 4.1, to each node, obtains its local density Rho.Rho is defined as the number of this node ' s length lower than the limit of certain predefine value Dc.
Step 4.2, to each node, obtains its dispersion Delta.Delta is defined as the length of side of most minor face in the adjacent side of all connections of this node higher Rho value neighbor node; If there is not such neighbor node, then get the length of side of the longest adjacent side of this node.
Rho value and Delta value respectively higher than the node of specific threshold R_T and D_T, are designated the Centroid of class by step 4.3.Each such node on behalf class, a namely visual human.
Other non-central nodes are classified as the shortest and Rho value of its distance that class higher than the Centroid of oneself by step 4.4.
Namely each node of step 4.5 same item represents and belongs to same visual human.Each class corresponding sets up corresponding visual human respectively
To clustering method shown in final steps 4, also other the conventional clustering methods as K-Means, hierarchical clustering (Hierarchical clustering) and so on can be adopted, they also can reach similar result, just different in complexity or effect.In conjunction with the clustering algorithm in this preferred embodiment, user behaviors log is analyzed, compare with other cluster such as K-Means, hierarchical clustering modes, reduce the complicated degree of analysis of whole system.Meanwhile, nationality is derived from the distribution characteristics amount of data itself by Delta and Rho value these two, and it is objective with reference to mode to provide the one that clusters number is selected.
In committed step 4.3, higher than certain respective threshold while that shown class central point identification method being Rho value and the Delta value of node.Other can be taked in reality based on the method for Rho value or Delta value.If Rho value is higher than 3, then delta value is between 4-5, and Rho value is higher than 5, then Delta value is between 5-6.
Below the implication of various value in visual human's method for building up of the present invention is described as follows in conjunction with simple examples.
The length of side characterizes: the measurement belonging to the possibility (similarity) of same person between node.
Rho characterizes: present node is to the importance of its abutment points.
Delta characterizes: if be class center with present node, the distinguishability at its other class centers relative.
For example:
The length of side may be defined as: two account numbers, in user behaviors log, work in coordination with the number of times (c occurred
a,b) inverse 1/ (c
a,b).I.e. two account numbers inverse of the successively movable number of times crossed in certain hour in same terminal.
Rho may be defined as: in the adjacent side of present node, and length is less than the quantity on the limit of parameter value Dc.
Delta is defined as the length of side of most minor face in the adjacent side of all connections of this node higher Rho value neighbor node; If there is not such neighbor node, then get the length of side of the longest adjacent side of this node.
Equation expression corresponding under above-mentioned definition example is:
Make c (a, b) for the collaborative occurrence number of the account number a that counts in subordinate act daily record and b, then have:
The length of side between 1.a, b:
D (a, b)=1/c (a, b) [equation 1].
2. all N number of neighbor node bn, n=1 item to a ... N (N is natural number), a's
Rho value:
Wherein, X (x) is defined as: if 1. x<0, then X (x)=1, otherwise X (x)=0.
The Delta value of 3.a:
The neighbor node of node a is made to be followed successively by b1 ... bN, then Delta (a) may be defined as:
1) if there is the adjacent side meeting Rho (bx) >Rho (a), then have:
Delta (a)=min{d (a, bn)) | n=1..N and Rho (bn) >Rho (a) }.
2) otherwise:
Delta(a)＝max{d(a,bn),n＝1..N}
Especially, for the node without any adjacent side, when marking its class mark, directly can be designated its oneself, namely independently forming a visual human.
Trying to achieve of Delta value is relevant with Rho value, and the definition of Rho value also can by other definition modes such as common centrads.
The data of value in practice with concrete of Dc are relevant, and we after obtaining connected graph, then can determine the value of Dc usually.That is, the same with in other common cluster modes, it is an input parameter.But with the K value in K-Means choose unlike, K value choose the number directly determining class, but Dc here can weaken the impact of subjective factor by the value of Rho value and Delta value and R_T and D_T because these parameters choose the objective consideration can introduced data self character.
A kind of method chosen of R_T and D_T is as follows.As shown in Figure 3, it is the Rho value-Delta Distribution value schematic diagram in visual human's method for building up one of the present invention preferred embodiment, and in figure, each point represents a node.First draw the Rho value-Delta Distribution value figure of each point, the distribution situation of observation Delta value (Rho value) afterwards, see that distribution situation there occurs sudden change, then getting this value is D_T (R_T) when which value.As in Fig. 3, at d ' (r ') place, the distribution situation of Delta value there occurs interruption/sudden change, then the value of D_T (R_T) is d ' (r ').If data point is more, then can sample, then do the reference of value with the distribution plan of sample point.
By introducing other models, such as attributes match, can using same for the matching result of corresponding model as the factor affecting edge lengths.That is, the factor introduced between account beyond the collaborative number of times occurred calculates the similarity between described account.
With attributes match for example, by mathematical symbolism, be namely as the parameter calculating the length of side using the result of attributes match.That is, the account number similarity of a and b making Match (a, b) arrive for attributes match, then can as the length of side of giving a definition:
d(a,b)＝f(c(a,b),match(a,b))。
For [equation 1], can select to be specifically defined as:
Introduce the length of side after attributes match model
See Fig. 4, it is the structural representation of visual human's apparatus for establishing one of the present invention preferred embodiment.Visual human's apparatus for establishing of this preferred embodiment comprises information extraction unit 1, connected graph tectonic element 2, and external model introduces unit 3, and visual human sets up unit 4 and visual human's merge cells 5.
Information extraction unit 1, for extracting account and the landing time corresponding with account, logging in end message in subordinate act daily record;
Connected graph tectonic element 2, for calculating the similarity between account according to situation about occurring collaborative between account, construct the connected graph characterizing account with node, and characterize the similarity between account with the length on the limit between node, limit between node is shorter, and between the account that node characterizes, similarity is higher;
External model introduces unit 3, calculates the similarity between described account for the factor introduced between account beyond collaborative situation about occurring;
Visual human sets up unit 4, for carrying out cluster to the node in described connected graph, sets up visual human according to cluster result;
Visual human's merge cells 5, for merging all visual humans and the account corresponding with visual human becomes Virtual Human Data storehouse.
Visual human's node set up in unit 4 pairs of connected graphs carry out cluster mode can with reference in aforementioned explanation to the description of visual human's method for building up of the present invention.
In visual human's method for building up of the present invention and device, by the mode of analytical behavior daily record, the result that actual analysis draws is " which account number belongs to same person operation ".In reality system demand, everyone is often more meaningful than account number for user, and this also can reduce because key values such as " ID (identity number) card No. " is untrue simultaneously, and causes the deviation in attribution of account numbers relational result.Analyze with user behaviors log, the Ke Shi – adding whole system only needs account number to identify, and might not need concrete account attributes.Be derived from the feature of user behaviors log and the reduction of above-mentioned complexity, the present invention can better be suitable for wider under, in longer time scope, the environment of more data amount.In fact, data acquisition from scope wider, the time is longer, data volume more conference make the actual accuracy rate of system higher.The present invention is according to above-mentioned to after user behaviors log analysis, and the attribution of account numbers relation that cluster draws, in conjunction with excessive datas such as account attributes, can depict the attribute informations such as the name of this visual human, address further.
In sum, visual human is set up in visual human's method for building up of the present invention and the daily record of device Behavior-based control, and complexity is low, and accuracy rate is high, is suitable for processing large data.
The above; for the person of ordinary skill of the art; can make other various corresponding change and distortion according to technical scheme of the present invention and technical conceive, and all these change and be out of shape the protection domain that all should belong to the accompanying claim of the present invention.
Claims (10)
1. visual human's method for building up, is characterized in that, comprises the steps:
Extract account and the landing time corresponding with account in subordinate act daily record, log in end message;
The similarity between account is calculated according to situation about occurring collaborative between account, construct the connected graph characterizing account with node, and characterizing the similarity between account with the length on the limit between node, the limit between node is shorter, and between the account that node characterizes, similarity is higher;
Cluster is carried out to the node in described connected graph, sets up visual human according to cluster result.
2. visual human's method for building up as claimed in claim 1, is characterized in that, the factor also can introduced between account beyond collaborative situation about occurring calculates the similarity between described account.
3. visual human's method for building up as claimed in claim 1, it is characterized in that, the process of the node in described connected graph being carried out to cluster comprises the steps:
Local density Rho, the Rho that obtain each node are respectively defined as the number of length lower than the adjacent side of predefine value Dc of this node of connection;
Obtain the dispersion Delta of each node respectively, Delta is defined as the length of side of most minor face in the adjacent side of all connections of this node higher Rho value neighbor node; If there is not such neighbor node, then get the length of side of the longest adjacent side of this node;
Be the Centroid of class higher than the node identification of predetermined threshold value R_T and D_T respectively by Rho value and Delta value;
Non-central node is classified as the shortest and Rho value of this non-central nodal distance higher than this non-central node Centroid belonging to class;
Each node of same item together forms a visual human.
4. visual human's method for building up as claimed in claim 1, is characterized in that, adopts K-Means method or hierarchy clustering method to carry out cluster to the node in described connected graph.
5. visual human's method for building up as claimed in claim 1, is characterized in that, also comprises merging all visual humans and the account corresponding with visual human becomes Virtual Human Data storehouse.
6. visual human's apparatus for establishing, is characterized in that, comprising:
Information extraction unit, for extracting account and the landing time corresponding with account, logging in end message in subordinate act daily record;
Connected graph tectonic element, for calculating the similarity between account according to situation about occurring collaborative between account, construct the connected graph characterizing account with node, and characterize the similarity between account with the length on the limit between node, limit between node is shorter, and between the account that node characterizes, similarity is higher;
Visual human sets up unit, for carrying out cluster to the node in described connected graph, sets up visual human according to cluster result.
7. visual human's apparatus for establishing as claimed in claim 6, is characterized in that, also comprises external model and introduces unit, calculates the similarity between described account for the factor introduced between account beyond collaborative situation about occurring.
8. visual human's apparatus for establishing as claimed in claim 6, it is characterized in that, the process of the node in described connected graph being carried out to cluster comprises the steps:
Local density Rho, the Rho that obtain each node are respectively defined as the number of length lower than the adjacent side of predefine value Dc of this node of connection;
Obtain the dispersion Delta of each node respectively, Delta is defined as the length of side of most minor face in the adjacent side of all connections of this node higher Rho value neighbor node; If there is not such neighbor node, then get the length of side of the longest adjacent side of this node;
Be the Centroid of class higher than the node identification of predetermined threshold value R_T and D_T respectively by Rho value and Delta value;
Non-central node is classified as the shortest and Rho value of this non-central nodal distance higher than this non-central node Centroid belonging to class;
Each node of same item together forms a visual human.
9. visual human's apparatus for establishing as claimed in claim 6, is characterized in that, adopts K-Means method or hierarchy clustering method to carry out cluster to the node in described connected graph.
10. visual human's apparatus for establishing as claimed in claim 6, is characterized in that, also comprise visual human's merge cells, for merging all visual humans and the account corresponding with visual human becomes Virtual Human Data storehouse.
Priority Applications (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410814330.4A CN104504264B (en) | 2014-12-08 | 2014-12-23 | Visual human's method for building up and device |
| PCT/CN2015/072487 WO2016090748A1 (en) | 2014-12-08 | 2015-02-09 | Virtual human creating method and apparatus |
Applications Claiming Priority (3)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2014107413344 | 2014-12-08 | ||
| CN201410741334 | 2014-12-08 | ||
| CN201410814330.4A CN104504264B (en) | 2014-12-08 | 2014-12-23 | Visual human's method for building up and device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN104504264A true CN104504264A (en) | 2015-04-08 |
| CN104504264B CN104504264B (en) | 2017-09-01 |
Family
ID=52945661
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201410814330.4A Active CN104504264B (en) | 2014-12-08 | 2014-12-23 | Visual human's method for building up and device |
Country Status (2)
| Country | Link |
|---|---|
| CN (1) | CN104504264B (en) |
| WO (1) | WO2016090748A1 (en) |
Cited By (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105224606A (en) * | 2015-09-02 | 2016-01-06 | 新浪网技术(中国)有限公司 | A kind of disposal route of user ID and device |
| WO2016106944A1 (en) * | 2014-12-31 | 2016-07-07 | 深圳市华傲数据技术有限公司 | Method for creating virtual human on mapreduce platform |
| CN105897667A (en) * | 2015-10-22 | 2016-08-24 | 乐视致新电子科技（天津）有限公司 | Device access history tracking method, apparatus, server and system |
| CN106372977A (en) * | 2015-07-23 | 2017-02-01 | 阿里巴巴集团控股有限公司 | Method and device for processing virtual account |
| WO2017028597A1 (en) * | 2015-08-20 | 2017-02-23 | 腾讯科技（深圳）有限公司 | Data processing method and apparatus for virtual resource |
| CN106604264A (en) * | 2017-01-04 | 2017-04-26 | 北京奇虎科技有限公司 | Application installation method and system, server, and mobile terminal |
| RU2617918C2 (en) * | 2015-06-19 | 2017-04-28 | Иосиф Исаакович Лившиц | Method to form person's image considering psychological portrait characteristics obtained under polygraph control |
| CN107248929A (en) * | 2017-05-27 | 2017-10-13 | 北京知道未来信息技术有限公司 | A kind of strong associated data generation method of multidimensional associated data |
| CN107291760A (en) * | 2016-04-05 | 2017-10-24 | 阿里巴巴集团控股有限公司 | Unsupervised feature selection approach, device |
| CN110032603A (en) * | 2019-01-22 | 2019-07-19 | 阿里巴巴集团控股有限公司 | The method and device that node in a kind of pair of relational network figure is clustered |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20090293121A1 (en) * | 2008-05-21 | 2009-11-26 | Bigus Joseph P | Deviation detection of usage patterns of computer resources |
| CN103544289A (en) * | 2013-10-28 | 2014-01-29 | 公安部第三研究所 | Feature extraction achieving method based on deploy and control data mining |
| CN103970752A (en) * | 2013-01-25 | 2014-08-06 | 北京思博途信息技术有限公司 | Estimating method and system for amount of unique visitors |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103368917B (en) * | 2012-04-01 | 2017-11-14 | 阿里巴巴集团控股有限公司 | A kind of risk control method and system of network virtual user |
| CN103927307B (en) * | 2013-01-11 | 2017-03-01 | 阿里巴巴集团控股有限公司 | A kind of method and apparatus of identification website user |
-
2014
- 2014-12-23 CN CN201410814330.4A patent/CN104504264B/en active Active
-
2015
- 2015-02-09 WO PCT/CN2015/072487 patent/WO2016090748A1/en active Application Filing
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20090293121A1 (en) * | 2008-05-21 | 2009-11-26 | Bigus Joseph P | Deviation detection of usage patterns of computer resources |
| CN103970752A (en) * | 2013-01-25 | 2014-08-06 | 北京思博途信息技术有限公司 | Estimating method and system for amount of unique visitors |
| CN103544289A (en) * | 2013-10-28 | 2014-01-29 | 公安部第三研究所 | Feature extraction achieving method based on deploy and control data mining |
Non-Patent Citations (1)
| Title |
|---|
| LAURENT GALLUCCIO 等: "《Graph based k-means clustering》", 《SIGNAL PROCESSING》 * |
Cited By (16)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2016106944A1 (en) * | 2014-12-31 | 2016-07-07 | 深圳市华傲数据技术有限公司 | Method for creating virtual human on mapreduce platform |
| RU2617918C2 (en) * | 2015-06-19 | 2017-04-28 | Иосиф Исаакович Лившиц | Method to form person's image considering psychological portrait characteristics obtained under polygraph control |
| CN106372977B (en) * | 2015-07-23 | 2019-06-07 | 阿里巴巴集团控股有限公司 | A kind of processing method and equipment of virtual account |
| CN106372977A (en) * | 2015-07-23 | 2017-02-01 | 阿里巴巴集团控股有限公司 | Method and device for processing virtual account |
| WO2017028597A1 (en) * | 2015-08-20 | 2017-02-23 | 腾讯科技（深圳）有限公司 | Data processing method and apparatus for virtual resource |
| CN106469413A (en) * | 2015-08-20 | 2017-03-01 | 深圳市腾讯计算机系统有限公司 | A kind of data processing method of virtual resource and device |
| CN106469413B (en) * | 2015-08-20 | 2021-08-03 | 深圳市腾讯计算机系统有限公司 | Data processing method and device for virtual resources |
| US10942949B2 (en) | 2015-08-20 | 2021-03-09 | Tencent Technology (Shenzhen) Company Limited | Data processing method and apparatus for virtual resource |
| CN105224606A (en) * | 2015-09-02 | 2016-01-06 | 新浪网技术(中国)有限公司 | A kind of disposal route of user ID and device |
| CN105224606B (en) * | 2015-09-02 | 2019-04-02 | 新浪网技术(中国)有限公司 | A kind of processing method and processing device of user identifier |
| CN105897667A (en) * | 2015-10-22 | 2016-08-24 | 乐视致新电子科技（天津）有限公司 | Device access history tracking method, apparatus, server and system |
| CN107291760A (en) * | 2016-04-05 | 2017-10-24 | 阿里巴巴集团控股有限公司 | Unsupervised feature selection approach, device |
| CN106604264A (en) * | 2017-01-04 | 2017-04-26 | 北京奇虎科技有限公司 | Application installation method and system, server, and mobile terminal |
| CN107248929A (en) * | 2017-05-27 | 2017-10-13 | 北京知道未来信息技术有限公司 | A kind of strong associated data generation method of multidimensional associated data |
| CN107248929B (en) * | 2017-05-27 | 2020-08-11 | 北京知道未来信息技术有限公司 | Strong correlation data generation method of multi-dimensional correlation data |
| CN110032603A (en) * | 2019-01-22 | 2019-07-19 | 阿里巴巴集团控股有限公司 | The method and device that node in a kind of pair of relational network figure is clustered |
Also Published As
| Publication number | Publication date |
|---|---|
| CN104504264B (en) | 2017-09-01 |
| WO2016090748A1 (en) | 2016-06-16 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104504264A (en) | Virtual person building method and device | |
| CN107281755B (en) | Detection model construction method and device, storage medium and terminal | |
| CN110413707A (en) | The excavation of clique's relationship is cheated in internet and checks method and its system | |
| CN105335400B (en) | Enquirement for user is intended to obtain the method and device of answer information | |
| CN104717124B (en) | A kind of friend recommendation method, apparatus and server | |
| Zhao et al. | A machine learning based trust evaluation framework for online social networks | |
| CN103795613A (en) | Method for predicting friend relationships in online social network | |
| CN106713290B (en) | Method for identifying main user account and server | |
| CN108022171B (en) | Data processing method and equipment | |
| CN113221104B (en) | Detection method of abnormal behavior of user and training method of user behavior reconstruction model | |
| CN110460458A (en) | Based on multistage markovian Traffic anomaly detection method | |
| CN110648172B (en) | Identity recognition method and system integrating multiple mobile devices | |
| CN110991789B (en) | Method and device for determining confidence interval, storage medium and electronic device | |
| CN104965846A (en) | Virtual human establishing method on MapReduce platform | |
| CN105376223A (en) | Network identity relationship reliability calculation method | |
| CN109639478A (en) | There are the method, apparatus of family relationship client, equipment and media for identification | |
| CN108764369A (en) | Character recognition method, device based on data fusion and computer storage media | |
| CN115883187A (en) | Method, device, equipment and medium for identifying abnormal information in network traffic data | |
| CN111241502A (en) | Cross-device user identification method and device, electronic device and storage medium | |
| CN109478219A (en) | For showing the user interface of network analysis | |
| CN107222319B (en) | Communication operation analysis method and device | |
| CN117172795A (en) | Intelligent technical service fee online consultation system | |
| CN115329078B (en) | Text data processing method, device, equipment and storage medium | |
| CN112052399A (en) | Data processing method and device and computer readable storage medium | |
| KR101928822B1 (en) | System and method for computing a user's trust value of unknown device in IoT |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| EXSB | Decision made by sipo to initiate 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: 518057 2203/2204, Building 1, Huide Building, North Station Community, Minzhi Street, Longhua District, Shenzhen, Guangdong Province Patentee after: SHENZHEN AUDAQUE DATA TECHNOLOGY Ltd. Address before: 518057 Rooms 713, 715 and 716, 7/F, Software Building, No. 9, High-tech Middle Road, High-tech Zone, Nanshan District, Shenzhen, Guangdong Province Patentee before: SHENZHEN AUDAQUE DATA TECHNOLOGY Ltd. |
|
| CP02 | Change in the address of a patent holder |