CN108681936A - A kind of fraud clique recognition methods propagated based on modularity and balance label - Google Patents
A kind of fraud clique recognition methods propagated based on modularity and balance label Download PDFInfo
- Publication number
- CN108681936A CN108681936A CN201810382121.5A CN201810382121A CN108681936A CN 108681936 A CN108681936 A CN 108681936A CN 201810382121 A CN201810382121 A CN 201810382121A CN 108681936 A CN108681936 A CN 108681936A
- Authority
- CN
- China
- Prior art keywords
- community
- node
- fraud
- modularity
- belonging
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/06—Buying, selling or leasing transactions
- G06Q30/0601—Electronic shopping [e-shopping]
- G06Q30/0609—Buyer or seller confidence or verification
Landscapes
- Business, Economics & Management (AREA)
- Accounting & Taxation (AREA)
- Finance (AREA)
- Development Economics (AREA)
- Economics (AREA)
- Marketing (AREA)
- Strategic Management (AREA)
- Physics & Mathematics (AREA)
- General Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
- Data Exchanges In Wide-Area Networks (AREA)
Abstract
The invention discloses a kind of fraud clique recognition methods propagated based on modularity and balance label, including：Using the known fraud mark of ID feature combinations user itself, similarity two-by-two is calculated to all users, similarity matrix is established, associated diagram is established by similarity matrix；Community and hierarchical information belonging to each node are obtained to the figure operation Louvain algorithms of foundation；Using community, hierarchical information and the fraud mark belonging to each node as the initial community information of each node, operation balance label communication process obtains each node finally affiliated community, further according to whether Kyodo News Service zoning subnetwork is belonged to, fraud clique is divided according to the fraud mark obtained is propagated.The fraud clique recognition methods propagated based on modularity and balance label is applied to the anti-anti- fraud field of cheating and merchandise of application by the present invention for the first time, it is associated with collection of illustrative plates using information architectures such as transaction associations, comprehensive corporations modularity information, using balance label propagation algorithm detection fraud corporations, potential fraudulent trading is taken precautions against.
Description
Technical field
The invention belongs to merchandise it is counter cheat and apply anti-fraud field, more particularly to it is a kind of based on modularity and balance label
The fraud clique recognition methods of propagation
Background technology
Case is cheated with the explosive growth of business on the lines such as e-commerce, Third-party payment, on line to be also becoming increasingly rampant,
And it shows that gimmick is changeable, the diversified trend in field, how effectively, identifies fraud on the line frequently occurred in time
Have become a problem in the urgent need to address.Fraud detection method on traditional line, generally directed to every online trading
Or merchant entities modeling, the correlated characteristic for implementing structure Business Stream carry out fraud detection, this method is to itself feature of merchandising
Apparent fraud excellent, but clique's relevance of fraudulent trading behind is had ignored, for forging normal users
Clique's fraud recognition capability of information is poor.
Community discovery is that wherein have corporations or the sub-network of specific rule by identifying from complex network structures, into
And complex network is divided, find out a kind of technology that its interior joint potentially contacts rule.In the anti-fraud of transaction and application
In anti-fraud field, client can build the complex web for highlighting abnormal behaviour by transaction association information and application related information
Network carries out analysis mining to the network using community discovery technology, can effectively identify the fraud clique in network, prevent to take advantage of
The generation of swindleness behavior.
Invention content
In view of the above-mentioned deficiencies in the prior art, it is an object of the present invention to provide a kind of based on modularity and balance label propagation
Cheat clique's recognition methods.
The purpose of the present invention is achieved through the following technical solutions：It is a kind of based on modularity and balance label propagate
Clique's recognition methods is cheated, this approach includes the following steps：
Step 1 is instead cheated in transaction or is applied in anti-fraud scene, extraction ID features；
Step 2, using the ID features extracted from transaction data or request for data, mark is cheated in conjunction with known to user itself
Know, similarity two-by-two is calculated to all users (including fraud blacklist and normal users), similarity matrix is established, passes through the phase
Associated diagram is established like degree matrix；
Step 3 obtains community and hierarchical information belonging to each node to the associated diagram operation Louvain algorithms of foundation；
Step 4, the community using belonging to each node, hierarchical information and fraud mark are believed as the initial community of each node
Breath, operation balance label communication process obtain each node finally belonging to community, further according to whether belonging to Kyodo News Service zoning subnetting
Network divides fraud clique according to the fraud mark obtained is propagated.
Further, in the step 2, the ID features include card number, account number, ip and device-fingerprint.
Further, in the step 2, if the n feature of user i is Xi,1,Xi,2,Xi,3….Xi,n, user i and user j
Similarity wi,jDefinition can refer to practical business situation, it is recommended to use predicable, COS distance etc., it is optional to be defined as follows：
Predicable：
wi,j=∑ku(Xi,k,Xj,k) (k=1 ... n)/k
COS distance：
wi,j=Cos (Xi,Xj)
For m user, following similarity matrix is formed：
Further, in the step 2, the setting 0 of threshold value p will be less than in similarity matrix, not be built for 0 user node
Stile contacts, and builds graph structure, and the similarity between node is the weight on side.
Further, the step 3 includes：
(1) assume that each node in associated diagram each belongs to an independent community when initial；
(2) its all neighbor node is traversed successively to each node i in associated diagram, calculating assigns it to neighbor node
Modularity variation delta Q before and after affiliated community；And update module degree variation maximum value max Δs Q, max_j are maximum value max Δs
The corresponding neighbor nodes of Q are assigned to node i as max Δ Q ＞ 0 in the community where max_j, otherwise remain unchanged；
(3) step (2) is re-executed, until node-home community no longer changes；
(4) one supernode of node merger for belonging to the same community in associated diagram is reconfigured into network, surpassed at this time
The weight of node is converted by the weight on side between community's interior nodes, and the side right weight between supernode is converted again by the side right between community, real
The compression of existing associated diagram；
(5) step (1) is re-executed, until the modularity of the iterations or associated diagram that reach setting no longer changes, most
Community of each node belonging to each level is obtained eventually.
Further, the step 4 includes：
(1) each node initial home in associated diagram that step 2 is established is set as each level that Louvain algorithms obtain
Affiliated community obtains each node<Community id, probability value>Information, wherein probability are equal to 1/ (the affiliated community of node
Quantity), community id is made of the community belonging to level, level, fraud mark；
(2) to its all neighbor node of each node traverses, the corresponding probability of identical community id is summed up, is denoted as
B, bmaxAfter being summed it up for probability<Community id, probability value>The maximum value of middle probability；According to formulaIt is every in filtration correlation figure
The community id of a node, wherein q are adjustable parameter, and q value ranges are between [0,1]；
(3) each node is normalized<Community id, probability value>Information；
(4) step (2) is repeated, until reaching specified number of iterations；
(5) according to whether ownership Kyodo News Service zoning subnetwork, fraud clique is divided according to the fraud mark obtained is propagated.
Beneficial effects of the present invention：The identification side of fraud clique that the present invention will be propagated for the first time based on modularity and balance label
Method is applied to the anti-anti- fraud field of cheating and merchandise of application, the advantages of combining Louvain and balance label propagation algorithm, not only
The related information of seed fraud node is utilized, it is also contemplated that corporations' modularity optimal demand is finally identified with suspicious
The fraud clique of trading activity and application action, has preferable social structure and outstanding accuracy rate.In the anti-fraud of transaction and
The anti-fraud field of application, this method have great research significance and use value.
Description of the drawings
Fig. 1 is the associated diagram schematic diagram established using similarity matrix；
Fig. 2 is the community and hierarchical information schematic diagram belonging to each node for being obtained according to associated diagram；
Fig. 3 is the result after being propagated into row label with fraud label in conjunction with level.
Specific implementation mode
With reference to the accompanying drawings and examples, the specific implementation mode of the present invention is described in further detail, is implemented below
Example is not limited to the scope of the present invention for illustrating the present invention.
The step of fraud clique recognition methods proposed by the present invention propagated based on modularity and balance label, is as follows：
The features such as step 1, extraction card number, account number, ip and device-fingerprint；As shown in table 1
1. transaction feature table of table
Step 2, using the feature extracted from transaction data to all users (including fraud blacklist and normal users)
Similarity two-by-two is calculated, similarity matrix is established, associated diagram is established by the matrix, as shown in Figure 1, the circle generation wherein in figure
Table user node, wherein digital representation User ID, in figure in upper digital representation the calculated side right weight of similarity matrix.
Step 3 obtains community and hierarchical information belonging to each node to the associated diagram operation Louvain algorithms of foundation,
As shown in Fig. 2, a plurality of nodes in the upper left corner are divided into a corporations, 3 figure compressions have been carried out altogether.
Step 4, the community using belonging to each node, hierarchical information and fraud mark are believed as the initial community of each node
Breath, operation balance label communication process obtain each node finally belonging to community, further according to whether belonging to Kyodo News Service zoning subnetting
Network divides fraud clique, as shown in figure 3, having identified three communities, wherein upper left altogether according to the fraud mark obtained is propagated
The black community at angle is fraud clique, and the white community in the upper right corner is normal users community, and the grey community of lower section is the group that leaves a question open
Group.
Wherein step 2 is specifically implemented according to the following steps：
If user i is characterized as Xi,1,Xi,2,Xi,3…·Xi,n, the similarity of user i and user j, which defines, can refer to reality
Service conditions, it is recommended to use predicable, COS distance etc., it is optional to be defined as follows：
Predicable：
wi,j=∑ku(Xi,k,Xj,k) (k=1 ... .n)/k
COS distance：
wi,j=Cos (Xi,Xj)
For m user, following similarity matrix is formed：
The setting 0 of threshold value p will be less than in similarity matrix, the user node for 0 does not establish side contact, builds graph structure,
Similarity between node is the weight on side, as shown in Figure 1, the circle wherein in figure represents user node, wherein digital representation is used
Family ID, in figure in upper digital representation the calculated side right weight of similarity matrix.
Wherein step 3 is specifically implemented according to the following steps：
(1) assume that each node in associated diagram each belongs to an independent community when initial；
(2) its all neighbor node is traversed successively to each node i in associated diagram, calculating assigns it to neighbor node
Modularity variation delta Q before and after affiliated community；And update module degree variation maximum value max Δs Q, max_j are maximum value max Δs
The corresponding neighbor nodes of Q are assigned to node i as max Δ Q ＞ 0 in the community where max_j, otherwise remain unchanged；
(3) step (2) is re-executed, until node-home community no longer changes；
(4) one supernode of node merger for belonging to the same community in associated diagram is reconfigured into network, surpassed at this time
The weight of node is converted by the weight on side between community's interior nodes, and the side right weight between supernode is converted again by the side right between community, real
The compression of existing associated diagram；
(5) step (1) is re-executed, until the modularity of the iterations or associated diagram that reach setting no longer changes, most
Community of each node belonging to each level is obtained eventually, as shown in Fig. 2, a plurality of nodes in the upper left corner are divided into a society
Group has carried out 3 figure compressions altogether.
The difference of number of edges under number of edges and random case of the modularity by calculating corporations' internal node, to weigh a community
Network divides quality, its value range is [- 1/2,1], is defined as follows：
Wherein, Ai,jThe weight on side between node i and node j；ki=∑jAi,jIndicate all sides being connected with node i
The sum of weight；ciIndicate the community belonging to node i；Indicate the sum of the weight on all sides.
In formulaThe probability that node j is connected to any one node isNow
Node i has kiThe number of degrees, therefore be on the side of random case lower node i and j
Wherein Δ Q is defined as follows：
∑ in indicates side right weight of the node emphasis in community in formula and ∑ tot indicates the side right in incident community
The sum of weight, ki,inIndicate node i incidence community the cum rights number of degrees and.
Wherein step 4 is specifically implemented according to the following steps：
(1) each node initial home in associated diagram that step 2 is established is set as each level that Louvain algorithms obtain
Affiliated community obtains each node<Community id, probability value>Information, wherein probability are equal to 1/ (the affiliated community of node
Quantity), community id is made of the community belonging to level, level, fraud mark；
(2) to its all neighbor node of each node traverses, the corresponding probability of identical community id is summed up, is denoted as
B, bmaxAfter being summed it up for probability<Community id, probability value>The maximum value of middle probability；According to formulaIt is every in filtration correlation figure
The community id of a node, wherein q are adjustable parameter, and q value ranges are between [0,1]；
(3) each node is normalized<Community id, probability value>Information；
(4) step (2) is repeated, until reaching specified number of iterations；
(5) according to whether ownership Kyodo News Service zoning subnetwork, divides fraud clique, such as according to the fraud mark obtained is propagated
Shown in Fig. 3, three communities are had identified altogether, and the wherein black community in the upper left corner is fraud clique, the white community in the upper right corner
Grey community for normal users community, lower section is the clique that leaves a question open.
The present invention proposes a kind of fraud clique recognition methods propagated based on modularity and balance label, combines
Louvain and the advantages of balance label propagation algorithm, is not only utilized the related information of seed fraud node, it is also contemplated that corporations
The optimal demand of modularity finally identifies the fraud clique with suspicious trading activity and application action, has preferable society
Group structure and outstanding accuracy rate.In the anti-fraud of transaction and the anti-fraud field of application, this method have great research significance and
Use value.
Embodiment of above is merely to illustrate the present invention, and not limitation of the present invention, in relation to the common of technical field
Technical staff without departing from the spirit and scope of the present invention, can be with various changes can be made and modification, therefore owns
Equivalent technical solution also belongs to scope of the invention, and scope of patent protection of the invention should be defined by the claims.
Claims (6)
1. it is a kind of based on modularity and balance label propagate the recognition methods of fraud clique, which is characterized in that this method include with
Lower step：
Step 1 is instead cheated in transaction or is applied in anti-fraud scene, extraction ID features；
Step 2, using the ID features extracted from transaction data or request for data, mark is cheated in conjunction with known to user itself,
Similarity two-by-two is calculated to all users including fraud blacklist and normal users, establishes similarity matrix, it is similar by this
Degree matrix establishes associated diagram；
Step 3 obtains community and hierarchical information belonging to each node to the associated diagram operation Louvain algorithms of foundation；
Step 4, the community using belonging to each node, hierarchical information and fraud are identified as the initial community information of each node,
Operation balance label communication process obtain each node finally belonging to community, further according to whether belonging to Kyodo News Service zoning subnetwork,
Fraud clique is divided according to the fraud obtained mark is propagated.
2. a kind of fraud clique recognition methods propagated based on modularity and balance label according to claim 1, special
Sign is, in the step 2, the ID features include card number, account number, ip and device-fingerprint.
3. a kind of fraud clique recognition methods propagated based on modularity and balance label according to claim 1, special
Sign is, in the step 2, if the n feature of user i is Xi,1,Xi,2,Xi,3….Xi,n, the similarity of user i and user j is fixed
Justice can refer to practical business situation and use predicable or COS distance.
4. a kind of fraud clique recognition methods propagated based on modularity and balance label according to claim 1, special
Sign is, in the step 2, the setting 0 of threshold value p will be less than in similarity matrix, and p is adjustable parameter, p value ranges [0,
1] between, the user node for 0 does not establish side contact, builds graph structure, and the similarity between node is the weight on side.
5. a kind of fraud clique recognition methods propagated based on modularity and balance label according to claim 1, special
Sign is that the step 3 includes：
(1) assume that each node in associated diagram each belongs to an independent community when initial；
(2) its all neighbor node is traversed successively to each node i in associated diagram, calculating assigns it to belonging to neighbor node
Modularity variation delta Q before and after community；And update module degree variation maximum value max Δs Q, max_j are Q pairs of maximum value max Δs
The neighbor node answered is assigned to node i as max Δ Q ＞ 0 in the community where max_j, otherwise remains unchanged；
(3) step (2) is re-executed, until node-home community no longer changes；
(4) one supernode of node merger for belonging to the same community in associated diagram is reconfigured into network, at this time supernode
Weight converted by the weight on side between community's interior nodes, the side right weight between supernode is converted again by the side right between community, is realized and is closed
Join the compression of figure；
(5) step (1) is re-executed, until the modularity of the iterations or associated diagram that reach setting no longer changes, final
To community of each node belonging to each level.
6. a kind of fraud clique recognition methods propagated based on modularity and balance label according to claim 1, special
Sign is that the step 4 includes：
(1) each node initial home in associated diagram that step 2 is established is set as belonging to each level that Louvain algorithms obtain
Community, obtain each node<Community id, probability value>Information, wherein probability are equal to 1/ (the affiliated community's number of node
Amount), community id is made of the community belonging to level, level, fraud mark；
(2) to its all neighbor node of each node traverses, the corresponding probability of identical community id is summed up, b is denoted as,
bmaxAfter being summed it up for probability<Community id, probability value>The maximum value of middle probability；According to formulaIt is every in filtration correlation figure
The community id of a node, wherein q are adjustable parameter, and q value ranges are between [0,1]；
(3) each node is normalized<Community id, probability value>Information；
(4) step (2) is repeated, until reaching specified number of iterations；
(5) according to whether ownership Kyodo News Service zoning subnetwork, fraud clique is divided according to the fraud mark obtained is propagated.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810382121.5A CN108681936B (en) | 2018-04-26 | 2018-04-26 | Fraud group identification method based on modularity and balanced label propagation |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810382121.5A CN108681936B (en) | 2018-04-26 | 2018-04-26 | Fraud group identification method based on modularity and balanced label propagation |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN108681936A true CN108681936A (en) | 2018-10-19 |
| CN108681936B CN108681936B (en) | 2021-11-02 |
Family
ID=63802082
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810382121.5A Active CN108681936B (en) | 2018-04-26 | 2018-04-26 | Fraud group identification method based on modularity and balanced label propagation |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108681936B (en) |
Cited By (33)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109685132A (en) * | 2018-12-20 | 2019-04-26 | 四川新网银行股份有限公司 | Label transmission method based on bigraph (bipartite graph) |
| CN109754258A (en) * | 2018-12-24 | 2019-05-14 | 同济大学 | It is a kind of based on individual behavior modeling towards online trading fraud detection method |
| CN109784636A (en) * | 2018-12-13 | 2019-05-21 | 中国平安财产保险股份有限公司 | Fraudulent user recognition methods, device, computer equipment and storage medium |
| CN109816535A (en) * | 2018-12-13 | 2019-05-28 | 中国平安财产保险股份有限公司 | Cheat recognition methods, device, computer equipment and storage medium |
| CN109949046A (en) * | 2018-11-02 | 2019-06-28 | 阿里巴巴集团控股有限公司 | The recognition methods of risk clique and device |
| CN110020866A (en) * | 2019-01-22 | 2019-07-16 | 阿里巴巴集团控股有限公司 | A kind of training method of identification model, device and electronic equipment |
| CN110070364A (en) * | 2019-03-27 | 2019-07-30 | 北京三快在线科技有限公司 | Method and apparatus, storage medium based on the fraud of graph model detection clique |
| CN110135890A (en) * | 2019-04-15 | 2019-08-16 | 深圳壹账通智能科技有限公司 | The product data method for pushing and relevant device of knowledge based relation excavation |
| CN110232524A (en) * | 2019-06-14 | 2019-09-13 | 哈尔滨哈银消费金融有限责任公司 | Social networks cheats the construction method of model, antifraud method and apparatus |
| CN110245875A (en) * | 2019-06-21 | 2019-09-17 | 深圳前海微众银行股份有限公司 | Risk of fraud appraisal procedure, device, equipment and storage medium |
| CN110490730A (en) * | 2019-08-21 | 2019-11-22 | 北京顶象技术有限公司 | Abnormal fund Assembling Behavior detection method, device, equipment and storage medium |
| CN110717828A (en) * | 2019-09-09 | 2020-01-21 | 中国科学院计算技术研究所 | Abnormal account detection method and system based on frequent transaction mode |
| CN110766091A (en) * | 2019-10-31 | 2020-02-07 | 上海观安信息技术股份有限公司 | Method and system for identifying road loan partner |
| CN110929509A (en) * | 2019-10-16 | 2020-03-27 | 上海大学 | Louvain community discovery algorithm-based field event trigger word clustering method |
| CN111241297A (en) * | 2020-01-06 | 2020-06-05 | 广州拉卡拉信息技术有限公司 | Map data processing method and device based on label propagation algorithm |
| CN111325578A (en) * | 2020-02-20 | 2020-06-23 | 深圳市腾讯计算机系统有限公司 | Prediction model sample determination method, prediction model sample determination device, prediction model sample determination medium, and prediction model sample determination device |
| CN111325350A (en) * | 2020-02-19 | 2020-06-23 | 第四范式（北京）技术有限公司 | Suspicious tissue discovery system and method |
| CN111428217A (en) * | 2020-04-12 | 2020-07-17 | 中信银行股份有限公司 | Method and device for identifying cheat group, electronic equipment and computer readable storage medium |
| CN111523831A (en) * | 2020-07-03 | 2020-08-11 | 支付宝(杭州)信息技术有限公司 | Risk group identification method and device, storage medium and computer equipment |
| CN111581530A (en) * | 2020-05-07 | 2020-08-25 | 北京互金新融科技有限公司 | Community division method and device |
| CN111598713A (en) * | 2020-07-24 | 2020-08-28 | 北京淇瑀信息科技有限公司 | Cluster recognition method and device based on similarity weight updating and electronic equipment |
| CN111612038A (en) * | 2020-04-24 | 2020-09-01 | 平安直通咨询有限公司上海分公司 | Abnormal user detection method and device, storage medium and electronic equipment |
| CN111641517A (en) * | 2020-04-29 | 2020-09-08 | 深圳壹账通智能科技有限公司 | Community division method and device for homogeneous network, computer equipment and storage medium |
| WO2020177450A1 (en) * | 2019-03-06 | 2020-09-10 | 京东数字科技控股有限公司 | Information merging method, transaction query method and apparatus, computer and storage medium |
| CN112184299A (en) * | 2020-09-23 | 2021-01-05 | 中国建设银行股份有限公司 | Arbitrage user identification method, apparatus, electronic device and medium |
| CN112785320A (en) * | 2021-02-01 | 2021-05-11 | 北京互金新融科技有限公司 | Credit risk determination method and device, storage medium and electronic equipment |
| CN112910888A (en) * | 2021-01-29 | 2021-06-04 | 杭州迪普科技股份有限公司 | Illegal domain name registration group mining method and device |
| CN113033966A (en) * | 2021-03-03 | 2021-06-25 | 携程旅游信息技术（上海）有限公司 | Risk target identification method and device, electronic equipment and storage medium |
| CN113129010A (en) * | 2020-01-10 | 2021-07-16 | 联洋国融(北京)科技有限公司 | Fraud group mining system and method based on complex network model |
| CN113326178A (en) * | 2021-06-22 | 2021-08-31 | 北京奇艺世纪科技有限公司 | Abnormal account number propagation method and device, electronic equipment and storage medium |
| CN113344697A (en) * | 2021-06-29 | 2021-09-03 | 平安普惠企业管理有限公司 | Group partner detection method, device, equipment and medium |
| CN114820219A (en) * | 2022-05-23 | 2022-07-29 | 杭银消费金融股份有限公司 | Complex network-based cheating community identification method and system |
| CN117575782A (en) * | 2024-01-15 | 2024-02-20 | 杭银消费金融股份有限公司 | Leiden community discovery algorithm-based group fraud identification method |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20100185935A1 (en) * | 2009-01-21 | 2010-07-22 | Nec Laboratories America, Inc. | Systems and methods for community detection |
| CN103729475A (en) * | 2014-01-24 | 2014-04-16 | 福州大学 | Multi-label propagation discovery method of overlapping communities in social network |
| CN107292424A (en) * | 2017-06-01 | 2017-10-24 | 四川新网银行股份有限公司 | A kind of anti-fraud and credit risk forecast method based on complicated social networks |
| CN107392559A (en) * | 2016-05-16 | 2017-11-24 | 阿里巴巴集团控股有限公司 | Institutional framework determination methods and device |
| CN107590504A (en) * | 2017-07-31 | 2018-01-16 | 阿里巴巴集团控股有限公司 | Abnormal main body recognition methods and device, server |
| CN107730262A (en) * | 2017-10-23 | 2018-02-23 | 阿里巴巴集团控股有限公司 | One kind fraud recognition methods and device |
| CN107943879A (en) * | 2017-11-14 | 2018-04-20 | 上海维信荟智金融科技有限公司 | Fraud group detection method and system based on social networks |
-
2018
- 2018-04-26 CN CN201810382121.5A patent/CN108681936B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20100185935A1 (en) * | 2009-01-21 | 2010-07-22 | Nec Laboratories America, Inc. | Systems and methods for community detection |
| CN103729475A (en) * | 2014-01-24 | 2014-04-16 | 福州大学 | Multi-label propagation discovery method of overlapping communities in social network |
| CN107392559A (en) * | 2016-05-16 | 2017-11-24 | 阿里巴巴集团控股有限公司 | Institutional framework determination methods and device |
| CN107292424A (en) * | 2017-06-01 | 2017-10-24 | 四川新网银行股份有限公司 | A kind of anti-fraud and credit risk forecast method based on complicated social networks |
| CN107590504A (en) * | 2017-07-31 | 2018-01-16 | 阿里巴巴集团控股有限公司 | Abnormal main body recognition methods and device, server |
| CN107730262A (en) * | 2017-10-23 | 2018-02-23 | 阿里巴巴集团控股有限公司 | One kind fraud recognition methods and device |
| CN107943879A (en) * | 2017-11-14 | 2018-04-20 | 上海维信荟智金融科技有限公司 | Fraud group detection method and system based on social networks |
Non-Patent Citations (1)
| Title |
|---|
| ZHI-HAO WU 等: "Balanced Multi-Label Propagation for Overlapping Community Detection in Social Networks", 《JOURNAL OF COMPUTER SCIENCE AND TECHNOLOGY》 * |
Cited By (50)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109949046B (en) * | 2018-11-02 | 2023-06-09 | 创新先进技术有限公司 | Identification method and device for risk group partner |
| CN109949046A (en) * | 2018-11-02 | 2019-06-28 | 阿里巴巴集团控股有限公司 | The recognition methods of risk clique and device |
| CN109784636A (en) * | 2018-12-13 | 2019-05-21 | 中国平安财产保险股份有限公司 | Fraudulent user recognition methods, device, computer equipment and storage medium |
| CN109816535A (en) * | 2018-12-13 | 2019-05-28 | 中国平安财产保险股份有限公司 | Cheat recognition methods, device, computer equipment and storage medium |
| CN109685132B (en) * | 2018-12-20 | 2020-12-11 | 四川新网银行股份有限公司 | Label propagation method based on bipartite graph |
| CN109685132A (en) * | 2018-12-20 | 2019-04-26 | 四川新网银行股份有限公司 | Label transmission method based on bigraph (bipartite graph) |
| CN109754258A (en) * | 2018-12-24 | 2019-05-14 | 同济大学 | It is a kind of based on individual behavior modeling towards online trading fraud detection method |
| CN109754258B (en) * | 2018-12-24 | 2023-05-12 | 同济大学 | Online transaction fraud detection method based on individual behavior modeling |
| CN110020866A (en) * | 2019-01-22 | 2019-07-16 | 阿里巴巴集团控股有限公司 | A kind of training method of identification model, device and electronic equipment |
| CN110020866B (en) * | 2019-01-22 | 2023-06-13 | 创新先进技术有限公司 | Training method and device for recognition model and electronic equipment |
| CN111666346A (en) * | 2019-03-06 | 2020-09-15 | 京东数字科技控股有限公司 | Information merging method, transaction query method, device, computer and storage medium |
| WO2020177450A1 (en) * | 2019-03-06 | 2020-09-10 | 京东数字科技控股有限公司 | Information merging method, transaction query method and apparatus, computer and storage medium |
| CN110070364A (en) * | 2019-03-27 | 2019-07-30 | 北京三快在线科技有限公司 | Method and apparatus, storage medium based on the fraud of graph model detection clique |
| WO2020192184A1 (en) * | 2019-03-27 | 2020-10-01 | 北京三快在线科技有限公司 | Gang fraud detection based on graph model |
| CN110135890A (en) * | 2019-04-15 | 2019-08-16 | 深圳壹账通智能科技有限公司 | The product data method for pushing and relevant device of knowledge based relation excavation |
| CN110232524A (en) * | 2019-06-14 | 2019-09-13 | 哈尔滨哈银消费金融有限责任公司 | Social networks cheats the construction method of model, antifraud method and apparatus |
| CN110245875A (en) * | 2019-06-21 | 2019-09-17 | 深圳前海微众银行股份有限公司 | Risk of fraud appraisal procedure, device, equipment and storage medium |
| CN110490730A (en) * | 2019-08-21 | 2019-11-22 | 北京顶象技术有限公司 | Abnormal fund Assembling Behavior detection method, device, equipment and storage medium |
| CN110490730B (en) * | 2019-08-21 | 2022-07-26 | 北京顶象技术有限公司 | Abnormal fund aggregation behavior detection method, device, equipment and storage medium |
| CN110717828B (en) * | 2019-09-09 | 2022-09-30 | 中国科学院计算技术研究所 | Abnormal account detection method and system based on frequent transaction mode |
| CN110717828A (en) * | 2019-09-09 | 2020-01-21 | 中国科学院计算技术研究所 | Abnormal account detection method and system based on frequent transaction mode |
| CN110929509A (en) * | 2019-10-16 | 2020-03-27 | 上海大学 | Louvain community discovery algorithm-based field event trigger word clustering method |
| CN110929509B (en) * | 2019-10-16 | 2023-09-15 | 上海大学 | Domain event trigger word clustering method based on louvain community discovery algorithm |
| CN110766091A (en) * | 2019-10-31 | 2020-02-07 | 上海观安信息技术股份有限公司 | Method and system for identifying road loan partner |
| CN110766091B (en) * | 2019-10-31 | 2024-02-27 | 上海观安信息技术股份有限公司 | Method and system for identifying trepanning loan group partner |
| CN111241297A (en) * | 2020-01-06 | 2020-06-05 | 广州拉卡拉信息技术有限公司 | Map data processing method and device based on label propagation algorithm |
| CN111241297B (en) * | 2020-01-06 | 2024-05-14 | 广州拉卡拉信息技术有限公司 | Atlas data processing method and apparatus based on label propagation algorithm |
| CN113129010A (en) * | 2020-01-10 | 2021-07-16 | 联洋国融(北京)科技有限公司 | Fraud group mining system and method based on complex network model |
| CN111325350A (en) * | 2020-02-19 | 2020-06-23 | 第四范式（北京）技术有限公司 | Suspicious tissue discovery system and method |
| CN111325350B (en) * | 2020-02-19 | 2023-09-29 | 第四范式（北京）技术有限公司 | Suspicious tissue discovery system and method |
| CN111325578B (en) * | 2020-02-20 | 2023-10-31 | 深圳市腾讯计算机系统有限公司 | Sample determination method and device of prediction model, medium and equipment |
| CN111325578A (en) * | 2020-02-20 | 2020-06-23 | 深圳市腾讯计算机系统有限公司 | Prediction model sample determination method, prediction model sample determination device, prediction model sample determination medium, and prediction model sample determination device |
| CN111428217A (en) * | 2020-04-12 | 2020-07-17 | 中信银行股份有限公司 | Method and device for identifying cheat group, electronic equipment and computer readable storage medium |
| CN111612038B (en) * | 2020-04-24 | 2024-04-26 | 平安直通咨询有限公司上海分公司 | Abnormal user detection method and device, storage medium and electronic equipment |
| CN111612038A (en) * | 2020-04-24 | 2020-09-01 | 平安直通咨询有限公司上海分公司 | Abnormal user detection method and device, storage medium and electronic equipment |
| CN111641517A (en) * | 2020-04-29 | 2020-09-08 | 深圳壹账通智能科技有限公司 | Community division method and device for homogeneous network, computer equipment and storage medium |
| CN111581530A (en) * | 2020-05-07 | 2020-08-25 | 北京互金新融科技有限公司 | Community division method and device |
| CN111523831A (en) * | 2020-07-03 | 2020-08-11 | 支付宝(杭州)信息技术有限公司 | Risk group identification method and device, storage medium and computer equipment |
| CN111598713B (en) * | 2020-07-24 | 2021-12-14 | 北京淇瑀信息科技有限公司 | Cluster recognition method and device based on similarity weight updating and electronic equipment |
| CN111598713A (en) * | 2020-07-24 | 2020-08-28 | 北京淇瑀信息科技有限公司 | Cluster recognition method and device based on similarity weight updating and electronic equipment |
| CN112184299A (en) * | 2020-09-23 | 2021-01-05 | 中国建设银行股份有限公司 | Arbitrage user identification method, apparatus, electronic device and medium |
| CN112910888A (en) * | 2021-01-29 | 2021-06-04 | 杭州迪普科技股份有限公司 | Illegal domain name registration group mining method and device |
| CN112785320B (en) * | 2021-02-01 | 2023-09-19 | 北京互金新融科技有限公司 | Credit risk determination method and device, storage medium and electronic equipment |
| CN112785320A (en) * | 2021-02-01 | 2021-05-11 | 北京互金新融科技有限公司 | Credit risk determination method and device, storage medium and electronic equipment |
| CN113033966A (en) * | 2021-03-03 | 2021-06-25 | 携程旅游信息技术（上海）有限公司 | Risk target identification method and device, electronic equipment and storage medium |
| CN113326178A (en) * | 2021-06-22 | 2021-08-31 | 北京奇艺世纪科技有限公司 | Abnormal account number propagation method and device, electronic equipment and storage medium |
| CN113344697A (en) * | 2021-06-29 | 2021-09-03 | 平安普惠企业管理有限公司 | Group partner detection method, device, equipment and medium |
| CN114820219A (en) * | 2022-05-23 | 2022-07-29 | 杭银消费金融股份有限公司 | Complex network-based cheating community identification method and system |
| CN117575782A (en) * | 2024-01-15 | 2024-02-20 | 杭银消费金融股份有限公司 | Leiden community discovery algorithm-based group fraud identification method |
| CN117575782B (en) * | 2024-01-15 | 2024-05-07 | 杭银消费金融股份有限公司 | Leiden community discovery algorithm-based group fraud identification method |
Also Published As
| Publication number | Publication date |
|---|---|
| CN108681936B (en) | 2021-11-02 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108681936A (en) | A kind of fraud clique recognition methods propagated based on modularity and balance label | |
| CN107730262B (en) | Fraud identification method and device | |
| CN110413707A (en) | The excavation of clique's relationship is cheated in internet and checks method and its system | |
| WO2022121145A1 (en) | Ethereum phishing scam detection method and apparatus based on graph classification | |
| CN108492173A (en) | A kind of anti-Fraud Prediction method of credit card based on dual-mode network figure mining algorithm | |
| CN112418520B (en) | Credit card transaction risk prediction method based on federal learning | |
| CN109636599A (en) | License block chain secret protection and monitoring and managing method based on group ranking | |
| CN107730384A (en) | Art sales method and server, server end and system based on block chain | |
| CN108765179A (en) | A kind of credible social networks analysis method calculated based on figure | |
| CN111652732B (en) | Bit coin abnormal transaction entity identification method based on transaction graph matching | |
| CN111738817B (en) | Method and system for identifying risk community | |
| CN109754258A (en) | It is a kind of based on individual behavior modeling towards online trading fraud detection method | |
| CN110795807B (en) | Construction method of element abnormal structure detection model based on complex network | |
| CN108764943B (en) | Suspicious user monitoring and analyzing method based on fund transaction network | |
| CN113283902B (en) | Multichannel blockchain phishing node detection method based on graphic neural network | |
| Li et al. | Retracted: Design of multimedia blockchain privacy protection system based on distributed trusted communication | |
| CN111724174A (en) | Citizen credit point evaluation method applying Xgboost modeling | |
| CN114358912A (en) | Risk weight fusion anomaly detection method based on federal learning | |
| CN113159866A (en) | Method for building network user trust evaluation model in big data environment | |
| CN112788064B (en) | Encryption network abnormal flow detection method based on knowledge graph | |
| CN110210858A (en) | A kind of air control guard system design method based on intelligent terminal identification | |
| Cortes et al. | Communities of interest | |
| CN115953240A (en) | Method for identifying fraudulent user in network transaction | |
| KR102163655B1 (en) | Method for monitoring blockchain and apparatus for using the method | |
| CN110110954B (en) | Risk vertex identification method and device |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| CP01 | Change in the name or title of a patent holder |
Address after: Room ABCD, 17th floor, building D, Paradise Software Park, No.3 xidoumen Road, Xihu District, Hangzhou City, Zhejiang Province, 310012 Patentee after: Zhejiang Bangsheng Technology Co.,Ltd. Address before: Room ABCD, 17th floor, building D, Paradise Software Park, No.3 xidoumen Road, Xihu District, Hangzhou City, Zhejiang Province, 310012 Patentee before: ZHEJIANG BANGSUN TECHNOLOGY Co.,Ltd. |
|
| CP01 | Change in the name or title of a patent holder |