CN110457893B - Method and equipment for acquiring account group - Google Patents
Method and equipment for acquiring account group Download PDFInfo
- Publication number
- CN110457893B CN110457893B CN201910671478.XA CN201910671478A CN110457893B CN 110457893 B CN110457893 B CN 110457893B CN 201910671478 A CN201910671478 A CN 201910671478A CN 110457893 B CN110457893 B CN 110457893B
- Authority
- CN
- China
- Prior art keywords
- association
- accounts
- account
- degree
- determining
- 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.)
- Active
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F21/00—Security arrangements for protecting computers, components thereof, programs or data against unauthorised activity
- G06F21/30—Authentication, i.e. establishing the identity or authorisation of security principals
- G06F21/45—Structures or tools for the administration of authentication
-
- 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
- G06Q20/00—Payment architectures, schemes or protocols
- G06Q20/38—Payment protocols; Details thereof
- G06Q20/42—Confirmation, e.g. check or permission by the legal debtor of payment
Abstract
Devices, methods, and systems for obtaining account groups are provided. The device comprises an acquisition unit for acquiring account relation data of each account in a plurality of accounts; a first determining unit, configured to determine a first association degree between each account of the plurality of accounts based on the account relationship data of each account; a construction unit, configured to construct an undirected graph based on a first degree of association between the respective accounts, where nodes in the undirected graph represent accounts, each undirected edge in the undirected graph represents that two accounts connected by the undirected edge are associated with each other, and the first degree of association meets a predetermined requirement; and a second determining unit for determining an account group associated with one or more predetermined accounts through a label propagation algorithm based on the undirected graph.
Description
Technical Field
The present invention relates to the field of internet technology, and in particular, to identifying a specific account group among multiple accounts.
Background
With the continuous development of internet technology, it is applied to various fields. People can log in and browse webpages and the like by means of the Internet technology, and can realize fund payment, account transfer and the like, so that great convenience is brought to people's life.
However, the utility model brings convenience to the life of people and brings potential safety hazard. For example, some individuals may log in through unauthorized computer technology and impersonate the accounts of others. For electronic paymate, this poses a threat to the user's funds security. Currently, an impersonated account is identified by independently analyzing account data of a single account to identify abnormal operations therein.
There is a need to provide faster and more reliable identification of fraudulent accounts.
Disclosure of Invention
It would be desirable to provide a method and apparatus that can perform data mining in a preferred full-size account based on a particular account (e.g., an imposter account) that has been identified to obtain a particular account group (e.g., an imposter group), so that potential imposter partners can be inferred from the identified, e.g., imposter account, and the identification efficiency of the imposter account can be improved.
According to one aspect, an apparatus for obtaining an account group is provided. The device comprises an acquisition unit for acquiring account relationship data of each account in a plurality of accounts; a first determining unit, configured to determine a first association degree between each account of the plurality of accounts based on the account relationship data of each account; a construction unit, configured to construct an undirected graph based on a first degree of association between the respective accounts, where nodes in the undirected graph represent accounts, each undirected edge in the undirected graph represents that two accounts connected by the undirected edge are associated with each other, and the first degree of association meets a predetermined requirement; and a second determining unit for determining an account group associated with one or more predetermined accounts through a label propagation algorithm based on the undirected graph.
According to another aspect, there is provided a method for obtaining an account group, comprising obtaining account relationship data for each of a plurality of accounts; determining a first association degree between each account in the plurality of accounts based on the account relationship data of each account; constructing an undirected graph based on the first association degree between the accounts, wherein nodes in the undirected graph represent the accounts, each undirected edge in the undirected graph represents the association between two accounts connected by the undirected edge, and the first association degree of the undirected edge meets the preset requirement; and determining, based on the undirected graph, an account group associated with one or more predetermined accounts through a label propagation algorithm.
According to yet another aspect, there is provided a system for obtaining a group of accounts, the system comprising a memory storing computer program code; and a processor running the computer program code to perform the method according to various embodiments.
According to the embodiments of each aspect of the present disclosure, a plurality of accounts, preferably full accounts, are taken as samples, data representing association relationships between the accounts is determined according to account relationship data of each account, an undirected graph is constructed based on the association relationships between every two accounts in the plurality of accounts, and undirected edges are constructed based on the association relationships between the accounts meeting predetermined requirements. Therefore, the method and the system can find out the partner associated with the specific account on the basis of the determined specific account such as the fraudulent account, pay attention to the relation between the accounts, and can mine potential fraudulent partners on the basis of analysis on a single fraudulent account, so that the identification efficiency of the fraudulent account is improved.
Drawings
FIG. 1 illustrates a block diagram of an apparatus for acquiring account groups, according to one embodiment;
FIG. 2 illustrates a flow diagram of a method for obtaining account groups, according to one embodiment;
FIG. 3 illustrates a flowchart of a method of determining a first degree of association between individual accounts according to one embodiment;
FIG. 4 illustrates a flowchart of a method of determining a first degree of association between individual accounts according to another embodiment;
FIG. 5 illustrates a flow diagram of a process for determining a first degree of association and building undirected edges according to one embodiment;
FIG. 6 illustrates a flow chart of a process for determining a first degree of association and constructing undirected edges according to another embodiment.
Various aspects and features of the present invention are described with reference to the foregoing drawings. The same or similar reference numerals are generally used to designate the same components. The drawings described above are only schematic and are non-limiting. The dimensions, shapes, numbers, or appearance of the various elements in the above figures may vary without departing from the spirit of the invention, and are not limited to that shown in the drawings of the specification alone.
Detailed Description
The application of embodiments of the present invention will be described below with reference to an application that identifies imposition of a group, it being understood that the application of embodiments of the present invention is not so limited and should be applicable in any application scenario where mining on a specific account basis is required to obtain a group of accounts associated with a specific account.
Fig. 1 illustrates a block diagram of a device 10 for acquiring account groups, according to one embodiment. The device 10 comprises an acquisition unit 11, a first determination unit 12, a construction unit 13, a second determination unit 14 and optionally a storage unit 15, an output unit 16.
The acquiring unit 11 acquires account related data of a plurality of accounts on an internet platform such as an electronic payment platform, and particularly acquires account related data of each of a full-size account on the internet platform.
In general, the account related data of each account may include data representing characteristics of the account itself and data representing a relationship occurring between the account and other accounts. The characteristic data of the account may include registration information of the account, such as the user's actual name, a certificate number, an associated funding card, etc., and/or operating environment data of the account, such as WiFi, IP address, operating handset/computer identification (mac address), etc. The data representing account relationships may include funds related data between different accounts, such as whether or not a convenient transfer or payment function (e.g., a paymate's affinity) is enabled with other accounts and with whom such a function is enabled, transfer records, etc.
The acquiring unit 11 further acquires account relationship data of each of the plurality of accounts. There may be one or more associations between any two accounts, including, for example, one or more of an operator, an operating device, an operating environment, funds, etc. The account relationship data between any two accounts can represent one or more of the above-mentioned association relationships between the two accounts. For example, two account numbers share the same computer, share the same WiFi, or open a close payment, etc. Or, in one embodiment, the account relationship data of each account includes relationship data between the account and the same operator, the same operation device, the same operation environment, and/or the same funding card and/or relationship data representing a relationship between the account and other accounts. The relationship data representing the relationship between the account number and the other account number includes data capable of representing a direct correlation occurring between the two account numbers, such as establishment of a close relationship between the two account numbers, transfer of money, redness package, and the like. In addition to the data representing the direct association between the two accounts, the relationship data between one of the accounts and the other accounts can also be derived from the relationship data between the respective accounts and the same operator, the same operating device, the same operating environment, and/or the same funding card.
In a specific embodiment, the account relationship data should include all data capable of representing one or more of the above-mentioned association relationships between two accounts, where the data may be relationship data obtained directly from account related data, for example, data related to opening a convenient transfer or payment function between two accounts in the account related data, common fund card data, common operator data, common operation device data, common operation environment data, etc.; data derived from account related data may also be used, such as data representing the earliest and/or latest time of correlation of two accounts, data representing the frequency of correlation of two accounts, data representing trends in correlation of two accounts, and data representing the time span of correlation of two accounts.
The first determining unit 12 processes the acquired multiple accounts and account relationship data thereof, and determines a first association degree between each of the multiple accounts. Specifically, the first association degree between any two accounts can be determined based on account relationship data related to one or more association relationships between the two accounts.
In one embodiment, when at least one association relationship exists between the respective accounts, the total association degree determined based on the second association degrees corresponding to the various association relationships between the respective accounts can be determined as the first association degree. Specifically, the second association degree corresponding to each association relationship may be determined based on account relationship data of any two accounts for each association relationship, and then the total association degree may be determined based on a plurality of second association degrees corresponding to various association relationships between the two accounts.
This may be achieved by scoring each association between two accounts based on their respective account relationship data according to a predetermined rule, and then determining the total association of the two accounts based on the score for each association. In determining the total association, the score for each association may be weighted according to a predetermined rule. The predetermined rule may be stored in the storage unit 15. The first degree of association between two accounts may also be determined by averaging or weighted averaging the second degree of association for each association.
In further embodiments, although there may be more than one relationship between two accounts, some relationships may not be indicative of a strong correlation between the two accounts, and therefore it is meaningful to remove relationships that do not meet predetermined criteria when determining the degree of correlation between the two accounts.
In one case, only those association relationships between two accounts for which the second association degree for each association relationship is greater than a predetermined threshold are considered in determining the first association degree between the two accounts. The first determining unit 12 removes, for each of the at least one association relationship, an account association relationship having a second association degree not greater than a corresponding threshold value from all the association relationships of any two accounts among the plurality of accounts, and determines a first association degree between the two accounts based on the at least one association relationship between the two accounts after the removal processing and the corresponding second association degree.
For example, there are association relations R1, R2, R3, R4, R5 between two accounts, which respectively represent five association relations involving a shared device, a shared fund card, a shared person, a close payment, and a transfer, the first determining unit 12 first determines a corresponding second association degree for each association relation between any two accounts, obtains 5 second association degrees A1, A2, A3, A4, A5 for R1, R2, R3, R4, R5, respectively, and then compares these 5 second association degrees with corresponding thresholds, respectively, those larger than the thresholds will be reserved, such as A1, A3, and A5 are reserved, and then the first association degrees may be determined based on the association relation reserved between the two accounts and the corresponding second association degrees A1, A3, and A5 thereof, for example, A1, A3, and A5 may be averaged or weighted equally.
In another case, when determining the first association degree between two accounts, only the order of the second association degree for each association relationship between the two accounts in the corresponding second association degrees between each two accounts in the plurality of accounts is considered, and the second association degree ordered before the predetermined position is used for determining the first association degree between the two accounts. In one example, the first determining unit 12 ranks, for each of the at least one association, the association between the respective accounts based on the second association degree that the respective accounts have; and removing the account association relation sequenced after the preset position from the association relation among the sequenced accounts, and then determining the first association degree among the accounts based on at least one association relation among the accounts after the removal treatment and the corresponding second association degree.
For example, the first determination unit 12 first ranks the respective second degrees of association for the association relationship R1 between the respective accounts, and the association relationship and the second degree of association before the predetermined position are retained, so that after ranking for all the association relationships, the first degree of association between the respective accounts is determined based on all the retained association relationships and the second degrees of association.
Although the association relationship and the corresponding second association degree to be considered in determining the first association degree between the two accounts are described before the predetermined position with reference to the fact that the second association degree for one association relationship between the two accounts is greater than the predetermined threshold and the row among the plurality of corresponding second association degrees between the respective accounts, respectively. It is also contemplated to combine the two cases that the first degree of association between the respective accounts is determined based only on those association relationships and the corresponding second degree of association between the respective accounts, which are greater than a predetermined threshold value for each association relationship and are arranged before a predetermined position among the corresponding plurality of second degrees of association of the plurality of accounts.
In the above embodiment, the first degree of association between any two account numbers may be determined in consideration of the second degree of association for each association relationship therebetween. It is also possible that when it is determined that there is a strong correlation with respect to one correlation between two accounts, the first correlation is determined only based on the correlation with the strong correlation and the second correlation corresponding thereto, or the second correlation corresponding to the correlation with the strong correlation is determined as the first correlation between the two accounts, and when the second correlation satisfies a predetermined criterion, the first correlation is considered to satisfy a predetermined requirement, and an undirected edge may be constructed between the two accounts.
In this embodiment, when there are multiple association relationships between the account numbers, the first association degree determined by the first determining unit may include at least one second association degree for at least one association relationship, where each second association degree corresponds to one association relationship between the account numbers. Preferably, for each association, a second association is determined for the current association between each account, and when the determined second association meets a predetermined requirement, the first association is considered to meet the predetermined requirement, and the undirected edge can be constructed between the two corresponding accounts by the construction unit. In a specific embodiment, the determined second association degree meeting the predetermined requirement may also be regarded as the first association degree between the two account numbers. In this embodiment, the relation of the second degree of association to a predetermined threshold and/or a predetermined position in the ranking, respectively, may also be considered.
In one case, the predetermined requirement includes that a second association degree between any two accounts is greater than a first predetermined threshold, and the first determining unit 12 is configured to, for any two accounts in each account, perform the following process for at least one association cycle until the calculated second association degree meets the predetermined requirement or the association degree calculation is completed for all association relations: and determining a second association degree corresponding to the current association relationship between the two accounts based on the account relationship data of the two accounts. After determining the second degree of association meeting the requirements, the first degree of association for the two accounts may be updated using the second degree of association meeting the requirements, and if necessary, it may be further determined in the subsequent construction unit 13 whether the first degree of association meets further predetermined requirements.
For example, for the various association relations R1, R2, R3, R4, R5 between any two accounts, sequentially determining a second association degree corresponding to each association relation, comparing the determined second association degree with a corresponding threshold, and stopping determining the second association degree when the determined second association degree is greater than the predetermined threshold, where it is determined that the association relation between the current two accounts meets the predetermined requirement, the construction unit 13 may construct an undirected edge between the two accounts.
In another case, the predetermined requirement includes that the magnitude of one second degree of association between any two accounts is ordered before the predetermined position in the magnitude of the corresponding second degree of association between the respective accounts. In this case, the first determining unit 12 performs the first association degree determining process and the building unit performs the undirected graph building process in a loop as will be described below for each of the at least one association relationship until the processing is completed for all the association relationships. Specifically, the first determining unit 12 determines, based on account relationship data of each current account in a plurality of current account pairs in the plurality of accounts, a second association degree corresponding to the current association relationship between each current account pair; ordering the association relation between each current account pair based on the second association degree; determining a current account pair set with association relations ordered before a preset position from the association relations among the current account pairs; and then removing the current account number pairs in the determined current account number pair set from the plurality of current account number pairs to serve as a plurality of current account number pairs in the next circulation process. In this way, the association relationships between the respective account numbers whose association degree values are ordered after the predetermined position for each association relationship are excluded from the first determination unit. In this case, the construction unit 13 can construct an undirected edge directly between each current account pair in the determined set of current account pairs.
A first degree of association between two accounts is typically determined among the plurality of accounts, so as to construct an undirected edge, and therefore, a plurality of account pairs exist among the plurality of accounts, and the first degree of association between each account pair is determined. For example, the first determining unit 12 determines the second degree of association for one association between the plurality of current account numbers (a, b), (a, c), (b, c), so that 3 second degrees of association can be obtained, they are ranked, if only the second degree of association and the corresponding association ranked first and the second degree of association for the current association between (a, c) are retained first, the first determining unit 12 determines that the second degree of association between (a, c) satisfies the predetermined requirement, the constructing unit 13 constructs an undirected edge between (a, c) based on which, while the first determining unit 12 can remove (a, c) from the plurality of current account numbers, and thereafter makes a judgment as to whether the above-described predetermined requirement is satisfied based on only (a, b), (b, c) for the next association.
It is also contemplated that the screening of the plurality of current account pairs of the set of current account pairs in the construction unit 13 may be further performed based on a second degree of association between the respective account pairs meeting predetermined requirements and further predetermined requirements.
The construction unit 13 constructs an undirected graph based on the first degree of association between the respective account numbers. The first degree of association may be a first degree of association including a plurality of second degrees of association as described above. The undirected graph thus generated takes each related account as a node, and the edges constructed between the accounts related to each other are taken as undirected edges therein. Whether or not there are edges connected to each other between different accounts can indicate whether or not there is an association relationship between them.
Although in some of the above embodiments, it may be determined by the first determination unit 12 whether the second degree of association included in the first degree of association satisfies a predetermined requirement. It is also contemplated that the first determining unit 12 determines only the first degree of association including the second degree of association, and the constructing unit 13 performs the above-described processing of whether or not the predetermined requirement is satisfied.
The construction unit 13 is further capable of setting the first degree of association between any two accounts to weights for the corresponding undirected edges to generate a weighted undirected graph. In a specific embodiment, the second association degree included in the first association degree and satisfying the predetermined requirement can be set as a weight for the corresponding undirected edge.
The optional storage unit 15 stores the above-mentioned predefined rule data, thresholds, requirements and/or criteria, etc. The predefined rule data can include rule data representing that there is an association between two accounts and/or rule data representing a degree of association between two accounts. In a specific embodiment, the predefined rule data can relate to rule data if the association between two accounts is evaluated.
As an example, assuming that it is determined from account relationship data of two accounts, the earliest transfer between the two accounts occurs before 20 days, and the number of transfers within the last 20 days is 10, wherein the transfer times corresponding to the intervals of 1-5 days, 6-10 days, 10-15 days, and 16-20 days are 1, 2, 3, 4, respectively, the relationship data can be evaluated/scored according to predefined rule data to obtain a second degree of association for transfer association between the two accounts. As can be seen from the above relationship data, the number of transfers between the two accounts increases with time, and therefore, the association relationship of the two accounts can be determined to be enhanced, at which time the second association degree indicating the degree of correlation can be set to be larger, and vice versa.
Although described above with reference to an association involving transfer, it will be appreciated that other types of associations are equally applicable. And when multiple association relations are involved between two accounts, the first association degree for the two accounts can be determined according to the second association degree for each association relation.
In some cases, for example, when determining whether the two accounts are operated by the same operator, the same operating device, the same fund card, and/or the same operating environment, data representing association between the two accounts and the same operator, the same operating device, the same fund card, and/or the same operating environment can be first determined according to the account related data of each of the two accounts, respectively, and then a second association degree representing association between the two accounts is determined based on the determined data of each of the two accounts. Therefore, the association relation between different accounts and the same operator, operation equipment, fund card and/or operation environment can be converted into the association relation between different accounts, and different accounts sharing the same operator, operation equipment (mobile phone or computer), operation environment (WiFi) and/or fund card can be determined.
The obtaining unit 11 also obtains one or more specific account numbers of the plurality of account numbers, for example, an already determined fraudulent account number, which can be achieved by user input. The second determining unit 14 determines an account group related to one or more specific accounts based on the one or more specific accounts from the acquiring unit 11 and the undirected graph or the weighted undirected graph from the constructing unit 13 by a tag propagation algorithm. The tag propagation algorithm is a known machine learning algorithm, and its specific implementation is not described in detail. In actual operation, a maximum iteration number can be set for the tag propagation algorithm, and when the propagation reaches the maximum iteration number, the propagation is terminated, so that time and calculation resources can be saved.
The output unit 16 outputs the account groups determined from the second determination unit 14, preferably only the accounts other than the acquired specific account in the determined account groups.
It was mentioned above that the first association degree for two accounts can be determined according to the relationship data for each association relationship when there may be a plurality of association relationships between the two accounts, that is, the first association degree representing the degree of correlation between the two accounts is determined by comprehensively considering each association relationship, which is not limitative. If multiple association relations exist between the fraudulent accounts in the fraudulent party, and multiple association relations exist between certain normal accounts, the scheme of the embodiment comprehensively considers the multiple association relations to obtain the first association corresponding to the two accounts can make calculation more complex, in fact, the fraudulent accounts may only have high association degree in a certain aspect, and the comprehensive consideration may obtain smaller first association degree, which may affect the accuracy of the finally obtained account group. Therefore, the first association degree comprising a plurality of second association degrees is disclosed, and once one second association degree for one association relation is determined to meet the preset requirement, undirected edges are constructed between corresponding accounts without calculating the second association degrees for all association relations between two accounts.
The respective embodiments are described above mainly with reference to the acquisition unit 11, the first determination unit 12, the construction unit 13, the second determination unit 14, the storage unit 15, and the output unit 16. It will be appreciated that the functions of the various units described above can be distributed among different units, and thus the functions of the various units can be split/combined/modified/altered to achieve corresponding effects. For example the first and second determination units can be combined.
For example, in one embodiment, the storage unit 15 can be omitted or incorporated in the construction unit 13, in which case the construction unit 13 can store predefined rule data.
A flowchart 200 of a method of acquiring an account group according to one embodiment of the invention is described below with reference to fig. 2.
At 210, account relationship data for each of a plurality of accounts is obtained. One or more association relationships may exist between any two accounts, where the one or more association relationships include one or more of an operator, an operating device, an operating environment, and/or a relationship of funds, where the relationship of funds includes a relationship related to transfer between two accounts, a transaction, a fund card, and the like. For example, the account relationship data for each account includes relationship data between the account and the same operator, the same operating device, the same operating environment, and/or the same funding card and/or relationship data representing a relationship between the account and other accounts.
At 220, a first degree of association between individual ones of the plurality of accounts is determined based on the account relationship data for each account. A first degree of association between any two of the plurality of accounts is typically determined.
In one embodiment, when at least one association relationship exists between the accounts, the first association degree is a total association degree determined based on second association degrees corresponding to various association relationships between the accounts. Thus, at 220, first, based on the account relationship data of each account, determining a second association degree corresponding to each association relationship in at least one association relationship between the accounts; and then determining the first association degree between the accounts based on at least one association relation between the accounts and the corresponding second association degree. The determination of the first degree of association in this embodiment will be described in detail below with respect to fig. 3 and 4.
In another embodiment, when multiple association relations exist between the accounts, the first association degrees are determined to include at least one second association degree, and each second association degree in the at least one second association degrees corresponds to one association relation between the accounts. In this way, it is convenient to find a second degree of association between two accounts that the degree of association value meets the predetermined requirement, and a undirected edge is constructed between the two accounts based on only the second degree of association. Note that the first degree of association is defined herein to include at least one second degree of association, but rather than constructing an undirected edge based on the at least one second degree of association, an undirected edge is constructed based on the found second degree of association meeting the predetermined requirement by merely sequentially determining the at least one second degree of association. The relevance determination and undirected edge construction based on this embodiment will be described below with reference to fig. 5 and 6.
At 230, an undirected graph of at least some of the plurality of accounts is constructed based on the determined first plurality of degrees of association, in which the accounts serve as nodes, each undirected edge representing an association between two accounts connected by the undirected edge and having a first degree of association that meets a predetermined requirement.
At 240, one or more particular accounts in the plurality of accounts are obtained. It is also contemplated that the particular account number may be obtained at 210 such that the process of 240 is omitted.
At 250, an account group for the one or more particular accounts is determined by a label propagation algorithm based on the obtained one or more particular accounts and the constructed undirected graph.
At 260, the obtained account group is output, and in one embodiment, the entire obtained account group can be output. In another embodiment, the account numbers obtained in the account group other than the first or more specific account numbers can be output.
The flow chart 200 of a method of obtaining an account group is described above with reference to the processing of 210-260. Fig. 3 illustrates a flow chart of a method 300 of determining a first degree of association between individual accounts at 220, according to one embodiment.
At 310, determining a second association degree corresponding to each association relationship in at least one association relationship between the accounts based on the account relationship data of each account;
At 320, removing account association relationships with a second association degree not greater than a corresponding threshold value from the association relationships of the respective accounts for each of the at least one association relationship;
at 330, a first association degree between the accounts is determined based on at least one association relationship between the accounts after the removal processing and the corresponding second association degree.
Fig. 4 illustrates a flow chart of a method 400 of determining a first degree of association between individual accounts at 220 according to another embodiment.
At 410, a second association corresponding to each association in at least one association between the accounts is determined based on the account relationship data of each account.
At 420, each association relationship between the accounts and the corresponding second association degree are obtained; for each association, sorting a plurality of association relations for the association relation among the account numbers based on the second association degree.
At 430, account associations ordered after a predetermined position are removed from the associations between the ordered individual accounts.
At 440, a first degree of association between the accounts is determined based on at least one association between the accounts after the removal process and the corresponding second degree of association.
FIG. 5 illustrates a flowchart of a process 500 for determining a first degree of association and building an undirected graph at 220 and 230 according to one embodiment. In this embodiment, the loop performs the following processing for each two of the plurality of accounts. And determining second association degrees of each association relation between any two accounts on the basis of the two accounts, and constructing undirected edges as long as one second association degree is larger than a preset threshold value without first calculating various second association degrees among the accounts aiming at all the association relations.
At 510, any two of the plurality of accounts are obtained. At 520, an association that may exist between the two accounts is obtained as a current association, for example, one of the associations involving the operator, the operating device, the operating environment, and the funds may be selected. At 530, a second degree of association for the current association between the two accounts is determined based on the account relationship data for the two accounts. At 540, it is determined whether the determined second degree of association meets a predetermined requirement, which may be that the second degree of association is greater than a first predetermined threshold.
If it is determined at 540 that the second degree of association meets the predetermined requirement, then at 550 a undirected edge is constructed between the two accounts, and then a return 510 is made to obtain a new any two accounts, at least one of the two newly obtained accounts being different from the previous any two accounts.
If it is determined at 540 that the second association does not meet the predetermined requirement, determining at 560 whether a new association is available, if so, returning 520 to acquire another association different from the previous association as the current association, and repeating the processes of 530 and 540; if not, return 510 to obtain new any two accounts.
Thus, after the above processing for all the different account numbers in the plurality of account numbers is completed, the processing is ended, and the currently constructed undirected graph is output.
FIG. 6 illustrates a flowchart of a process 600 for determining a first degree of association and building an undirected graph at 220 and 230 according to one embodiment. In this embodiment, the predetermined requirement includes that the magnitude of one second degree of association between two accounts be ordered before the predetermined position by the magnitude in the corresponding second degree of association between the respective accounts. In this embodiment, based on each association relationship, first, each second association degree of each account with respect to one association relationship is obtained, the association degrees are ordered, an undirected edge can be constructed between corresponding accounts whose second association degree is arranged before a predetermined position, then account pairs with undirected edges constructed are excluded from a plurality of account pairs in the current plurality of accounts, each second association degree of each account pair with respect to the next association relationship is obtained based on a plurality of account pairs after the exclusion treatment, and the subsequent ordering, construction and exclusion treatment is repeated until the treatment is performed for each association relationship.
According to this embodiment, at 610, one of the association relationships that may exist between the respective account numbers is acquired as the current association relationship, and for example, one of the association relationships involving the operator, the operation device, the operation environment, and the funds may be selected. At 620, a current account pair is obtained, which includes removing those account pairs from the previous current account pair for which the previous association, the second association thereof, has met the predetermined requirement and an undirected edge is constructed, thereby obtaining a new current account pair. At 630, a second degree of association for the current association between each account pair is determined based on account relationship data for each current account in the current account pair. At 640, the plurality of second degrees of association obtained at 630 for the plurality of current account pairs, respectively, are ordered. At 650, a set of current account number pairs whose second degree of association is ranked before the predetermined location is determined, and an undirected edge is constructed between each current account number pair in the determined set of current account number pairs. At 660, it is determined whether there are new associations available. If it is determined at 660 that there are more, a new association is obtained as a current association by returning to 610, and each current account in the determined set of current account pairs arranged before the predetermined position is removed from the plurality of current account pairs at 620, thereby obtaining a new current account pair. The process of 630-650 is repeated for the new current account pair. If it is determined at 660 that no new associations are available, processing ends at 670, outputting the currently constructed undirected graph.
Although various embodiments of the method according to the present invention are described with reference to the flowcharts shown in fig. 2-6. It can be understood that corresponding processes can be added/modified/deleted on the basis of the flowcharts of the above embodiments, so as to constitute a new technical solution to achieve different effects; furthermore, the processes in the above-described different embodiments can be arbitrarily split up in combination to achieve the respective functions and effects.
For example, the embodiments shown in fig. 5 and 6 may incorporate, i.e., build undirected edges for only those account pairs whose second degree of association is greater than a corresponding threshold and which are arranged before a predetermined position in the ordering of corresponding second degrees of association for the same association for each account.
Or in a further embodiment, the first association degree between any two accounts/the second association degree included by the first association degree and meeting the preset requirement can be set as a weight for a corresponding undirected edge, so as to generate a weighted undirected graph, and an account group is determined through a label propagation algorithm based on the weighted undirected graph.
It will be appreciated that the functions of the various units in the device for account group and the flow of the method of the various embodiments of the invention can be implemented by a computer program/software. The software comprises computer program instructions capable of being loaded into a working memory of a data processor for performing a method according to embodiments of the invention when run.
Exemplary embodiments of the present invention cover both: the computer program/software of the invention is created/used from the beginning and the existing program/software is converted into the computer program/software of the invention by means of an update.
According to further embodiments of the present invention, a machine (e.g., computer) readable medium, such as a CD-ROM, is provided, wherein the readable medium has computer program code stored thereon, which when executed, causes a computer or processor to perform a method according to embodiments of the present invention. The machine-readable medium is, for example, an optical storage medium or a solid-state medium supplied together with or as part of other hardware.
The computer program for performing the methods according to embodiments of the invention may also be distributed in other forms, such as via the internet or other wired or wireless telecommunication systems. The computer program may also be provided on a network, such as the world wide web, and be downloaded into the working computers of the data processors from such a network.
It is also understood that the various units in the device for obtaining account groups and the flow of the method according to the embodiments of the present invention can also be implemented by hardware or a combination of hardware and software.
In one embodiment, a system for obtaining a group of accounts can be implemented by a memory and a processor. The memory is capable of storing computer program code for running a method flow according to various embodiments of the invention; when executing program code from memory, the processor performs the flow according to various embodiments of the invention.
The foregoing describes specific embodiments of the present disclosure. Other embodiments are within the scope of the following claims. In some cases, the actions or steps recited in the claims can be performed in a different order than in the embodiments and still achieve desirable results. In addition, the processes depicted in the accompanying figures do not necessarily require the particular order shown, or sequential order, to achieve desirable results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.
It has to be noted that embodiments of the invention are described with reference to different subject-matters. In particular, some embodiments are described with reference to method-type claims, while other embodiments are described with reference to apparatus-type claims. However, one skilled in the art will recognize from the above and following description that, unless otherwise indicated, any combination of features relating to different subject matter is also considered to be disclosed herein, in addition to any combination of features belonging to one type of subject matter. And, all features can be combined, providing a synergistic effect that is greater than the simple addition of features.
The present invention has been described above with reference to specific embodiments, and it will be understood by those skilled in the art that the technical solution of the present invention can be implemented in various ways without departing from the spirit and essential characteristics of the present invention. The particular embodiments are illustrative only and not limiting. In addition, the embodiments can be arbitrarily combined to achieve the object of the present invention. The scope of the invention is defined by the appended claims.
The word "comprising" in the description and claims does not exclude the presence of other elements or steps. The functions of the respective elements described in the specification or recited in the claims may also be divided or combined, and implemented by corresponding plural elements or single element.
Claims (23)
1. An apparatus for obtaining account groups, comprising
The system comprises an acquisition unit, a storage unit and a storage unit, wherein the acquisition unit is used for acquiring account relation data of each account in a plurality of accounts, and the plurality of accounts are full accounts of an Internet platform;
a first determining unit, configured to determine a first association degree between each account of the plurality of accounts based on the account relationship data of each account;
a construction unit, configured to construct an undirected graph based on a first degree of association between the respective accounts, where nodes in the undirected graph represent accounts, each undirected edge in the undirected graph represents that two accounts connected by the undirected edge are associated with each other, and the first degree of association meets a predetermined requirement; and
A second determining unit, configured to determine, by using a tag propagation algorithm, an account group associated with one or more predetermined accounts based on the undirected graph and the one or more predetermined accounts, where the one or more predetermined accounts are determined fraudulent accounts.
2. The device of claim 1, wherein the first degree of association is a total degree of association determined based on a second degree of association corresponding to each of the relationships between the respective accounts when at least one relationship exists between the respective accounts.
3. The apparatus of claim 1, wherein the first degree of association includes at least one second degree of association when there are a plurality of relationships between the respective accounts, each of the at least one second degree of association corresponding to one relationship between the respective accounts.
4. The apparatus of claim 2, wherein the first determining unit is configured to:
based on the account relationship data of each account, determining a second association degree corresponding to each association relationship in the at least one association relationship among the accounts; and
and determining the first association degree between the accounts based on the at least one association relation between the accounts and the corresponding second association degree.
5. The apparatus of claim 4, wherein the first determining unit is further configured to:
removing account association relationships having a second association degree not greater than a corresponding threshold value from the association relationships of the respective accounts for each association relationship of the at least one association relationship, and
and determining the first association degree between the account numbers based on at least one association relation between the account numbers after the removal processing and the corresponding second association degree.
6. The apparatus of claim 4, wherein the first determining unit is further configured to:
ordering the association relationships among the account numbers based on the second association degree for each association relationship in the at least one association relationship;
removing the account association relations after being ordered at the preset positions from the association relations among the ordered accounts; and
and determining the first association degree between the account numbers based on at least one association relation between the account numbers after the removal processing and the corresponding second association degree.
7. A device according to claim 3, wherein the predetermined requirement comprises a second degree of association between the two account numbers being greater than a first predetermined threshold, the first determining unit being configured to:
For any two accounts in each account, for the at least one association, the following processes are circularly executed until the calculated second association meets the predetermined requirement or association calculation is completed for all the association: and determining a second association degree corresponding to the current association relationship between the two accounts based on the account relationship data of the two accounts.
8. The device of claim 3, wherein the predetermined requirement includes that a magnitude of one second degree of association between the two accounts is ordered before a predetermined position in a magnitude of a corresponding second degree of association between the respective accounts,
for each of the at least one association, the first determining unit circularly performs a first association determining process, and the building unit circularly performs an undirected graph building process until processing is completed for all of the association,
wherein the first determining unit is configured to: determining a second association degree corresponding to the current association relationship between each current account pair based on account relationship data of each current account in a plurality of current account pairs in the plurality of accounts;
Ranking the association between each current account pair based on the second association degree;
determining a current account pair set with association relations ordered before a preset position from the association relations among the current account pairs; and
removing the current account number pair in the determined current account number pair set from the plurality of current account number pairs to serve as a plurality of current account number pairs in the next circulation process,
the construction unit is used for:
and constructing undirected edges between each current account pair in the determined current account pair set.
9. The apparatus according to any one of claims 1-7, wherein the construction unit sets a first degree of association between any two accounts as a weight for a corresponding undirected edge to generate a weighted undirected graph, and the second determination unit determines the account group by a tag propagation algorithm based on the weighted undirected graph.
10. The device of any of claims 2-7, wherein the at least one association includes one or more of an operator, an operating device, an operating environment, and/or a funding association.
11. The device of claim 10, wherein the account relationship data for each account includes relationship data between the account and the same operator, the same operating device, the same operating environment, and/or the same funding card and/or relationship data representing a relationship between the account and other accounts.
12. A method for acquiring account groups comprises
Acquiring account relationship data of each account in a plurality of accounts, wherein the plurality of accounts are full accounts of an Internet platform;
determining a first association degree between each account in the plurality of accounts based on the account relationship data of each account;
constructing an undirected graph based on the first association degree between the accounts, wherein nodes in the undirected graph represent the accounts, each undirected edge in the undirected graph represents the association between two accounts connected by the undirected edge, and the first association degree of the undirected edge meets the preset requirement; and
an account group associated with one or more predetermined accounts is determined by a tag propagation algorithm based on the undirected graph and the one or more predetermined accounts, the one or more predetermined accounts being determined fraudulent accounts.
13. The method of claim 12, wherein the first degree of association is a total degree of association determined based on a second degree of association corresponding to each of the relationships between the respective accounts when at least one relationship exists between the respective accounts.
14. The method of claim 12, wherein when there are multiple associations between the respective accounts, the first association includes at least one second association, each of the at least one second association corresponding to one of the associations between the respective accounts.
15. The method of claim 13, wherein determining a first degree of association between the respective accounts based on the account relationship data for each account comprises:
based on the account relationship data of each account, determining a second association degree corresponding to each association relationship in at least one association relationship among the accounts; and
and determining the first association degree between the accounts based on at least one association relation between the accounts and the corresponding second association degree.
16. The method of claim 15, wherein determining a first degree of association between the respective accounts based on the account relationship data for each account further comprises:
for each of the at least one association,
removing account association relations with a second association degree not greater than a corresponding threshold value from the association relations of the accounts, and
based on at least one association relationship between the accounts and the corresponding second association degree, determining the first association degree between the accounts comprises:
and determining the first association degree between the account numbers based on at least one association relation between the account numbers after the removal processing and the corresponding second association degree.
17. The method of claim 15, wherein determining a first degree of association between the individual accounts based on the account relationship data for each account further comprises:
for each of the at least one association,
ranking the association relationships between the respective accounts based on the second degree of association;
removing the account association relation ordered after a preset position from the association relation among the ordered accounts, and
based on at least one association relationship between the accounts and the corresponding second association degree, determining the first association degree between the accounts comprises:
and determining the first association degree between the account numbers based on at least one association relation between the account numbers after the removal processing and the corresponding second association degree.
18. The method of claim 14, wherein the predetermined requirement includes a second degree of association between the two accounts being greater than a first predetermined threshold, and determining the first degree of association between the respective accounts based on the account relationship data for each account comprises:
for any two accounts in each account, for the at least one association, the following processes are circularly executed until the calculated second association meets the predetermined requirement or association calculation is completed for all the association:
And determining a second association degree corresponding to the current association relationship between the two accounts based on the account relationship data of the two accounts.
19. The method of claim 14, wherein the predetermined requirement includes a magnitude of one second degree of association between the two accounts ordered before a magnitude in a corresponding second degree of association between the respective accounts,
the method comprises the following steps: for each of the at least one association, performing a first association determination process and an undirected graph construction process in a loop until processing is completed for all of the associations,
the first association degree determining process includes:
determining a second association degree corresponding to the current association relationship between each current account pair based on account relationship data of each current account in a plurality of current account pairs in the plurality of accounts;
ranking the association between each current account pair based on the second association degree;
determining a current account pair set with association relations ordered at a preset position from the association relations among the current account pairs; and
removing the current account number pair in the determined current account number pair set from the plurality of current account number pairs to serve as a plurality of current account number pairs in the next circulation process,
The undirected graph construction process comprises the following steps:
and constructing undirected edges between each current account pair in the determined current account pair set.
20. The method of any of claims 12-19, wherein constructing the undirected graph based on a first degree of association between any two accounts further comprises:
a first degree of association between any two accounts is set as a weight for the corresponding undirected edge to generate a weighted undirected graph,
and wherein the account group is determined by a tag propagation algorithm based on the weighted undirected graph.
21. The method of any of claims 13-19, wherein the at least one association comprises one or more of an operator, an operating device, an operating environment, and/or a funding association.
22. The method of claim 21, wherein the account relationship data for each account includes relationship data between the account and the same operator, the same operating device, the same operating environment, and/or the same funding card and/or relationship data representing a relationship between the account and other accounts.
23. A system for acquiring account groups comprises
A memory storing computer program code; and
a processor running the computer program code to perform the method according to any one of claims 12-22.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910671478.XA CN110457893B (en) | 2019-07-24 | 2019-07-24 | Method and equipment for acquiring account group |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910671478.XA CN110457893B (en) | 2019-07-24 | 2019-07-24 | Method and equipment for acquiring account group |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110457893A CN110457893A (en) | 2019-11-15 |
| CN110457893B true CN110457893B (en) | 2023-05-05 |
Family
ID=68483281
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910671478.XA Active CN110457893B (en) | 2019-07-24 | 2019-07-24 | Method and equipment for acquiring account group |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110457893B (en) |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109102151A (en) * | 2018-07-03 | 2018-12-28 | 阿里巴巴集团控股有限公司 | A kind of suspicious group identification method and apparatus |
| CN109242515A (en) * | 2018-08-29 | 2019-01-18 | 阿里巴巴集团控股有限公司 | Cross-platform abnormal account recognition methods and device |
| CN109598509A (en) * | 2018-10-17 | 2019-04-09 | 阿里巴巴集团控股有限公司 | The recognition methods of risk clique and device |
| CN109902698A (en) * | 2017-12-08 | 2019-06-18 | 北京京东尚科信息技术有限公司 | Information generating method and device |
Family Cites Families (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US10290053B2 (en) * | 2009-06-12 | 2019-05-14 | Guardian Analytics, Inc. | Fraud detection and analysis |
| US20140303993A1 (en) * | 2013-04-08 | 2014-10-09 | Unisys Corporation | Systems and methods for identifying fraud in transactions committed by a cohort of fraudsters |
| US9396472B2 (en) * | 2014-10-31 | 2016-07-19 | Facebook, Inc. | Systems and methods for dynamically identifying illegitimate accounts based on rules |
| CN109271418B (en) * | 2018-08-14 | 2022-03-04 | 蚂蚁智安安全技术(上海)有限公司 | Suspicious group identification method, device, equipment and computer readable storage medium |
| CN109255632A (en) * | 2018-09-03 | 2019-01-22 | 武汉斗鱼网络科技有限公司 | A kind of user community recognition methods, device, equipment and medium |
| 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 |
| CN110046929B (en) * | 2019-03-12 | 2023-06-20 | 平安科技（深圳）有限公司 | Fraudulent party identification method and device, readable storage medium and terminal equipment |
| CN110009216A (en) * | 2019-03-28 | 2019-07-12 | 阿里巴巴集团控股有限公司 | Credit risk subscribers recognition methods and device |
-
2019
- 2019-07-24 CN CN201910671478.XA patent/CN110457893B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109902698A (en) * | 2017-12-08 | 2019-06-18 | 北京京东尚科信息技术有限公司 | Information generating method and device |
| CN109102151A (en) * | 2018-07-03 | 2018-12-28 | 阿里巴巴集团控股有限公司 | A kind of suspicious group identification method and apparatus |
| CN109242515A (en) * | 2018-08-29 | 2019-01-18 | 阿里巴巴集团控股有限公司 | Cross-platform abnormal account recognition methods and device |
| CN109598509A (en) * | 2018-10-17 | 2019-04-09 | 阿里巴巴集团控股有限公司 | The recognition methods of risk clique and device |
Non-Patent Citations (2)
| Title |
|---|
| Tianqing Zhu.Suspicious Financial Transaction Detection Based on Empirical Mode Decomposition Method.2006 IEEE Asia-Pacific Conference on Services Computing (APSCC'06).2006,第300-304页. * |
| 潘理 ; 吴鹏 ; 黄丹华 ; .在线社交网络群体发现研究进展.电子与信息学报.2017,(第09期),全文. * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110457893A (en) | 2019-11-15 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109598509B (en) | Identification method and device for risk group partner | |
| CN107193894B (en) | Data processing method, individual identification method and related device | |
| US8694980B2 (en) | Efficient egonet computation in a weighted directed graph | |
| US10282542B2 (en) | Information processing apparatus, information processing method, and computer readable medium | |
| CN112543176A (en) | Abnormal network access detection method, device, storage medium and terminal | |
| CN106682906B (en) | Risk identification and service processing method and equipment | |
| CN106909543B (en) | Pattern matching method and device of rule engine | |
| KR20110131094A (en) | Identifying communities in an information network | |
| CN111046237B (en) | User behavior data processing method and device, electronic equipment and readable medium | |
| CN106921504B (en) | Method and equipment for determining associated paths of different users | |
| CN110881050A (en) | Security threat detection method and related product | |
| KR102086936B1 (en) | User data sharing method and device | |
| CN109460930B (en) | Method for determining risk account and related equipment | |
| Ashraf et al. | WeFreS: weighted frequent subgraph mining in a single large graph | |
| CN111127185A (en) | Credit fraud identification model construction method and device | |
| CN113689270A (en) | Method for determining black product device, electronic device, storage medium, and program product | |
| CN113065748A (en) | Business risk assessment method, device, equipment and storage medium | |
| CN110457893B (en) | Method and equipment for acquiring account group | |
| CN102790707A (en) | Method and device for classifying object | |
| CN104994179B (en) | A kind of data processing method and server | |
| WO2019091101A1 (en) | Salary calculation method, application server and computer-readable storage medium | |
| CN109992960A (en) | A kind of forgery parameter detection method, device, electronic equipment and storage medium | |
| CN109995834A (en) | Massive dataflow processing method, calculates equipment and storage medium at device | |
| CN114781517A (en) | Risk identification method and device and terminal equipment | |
| CN113553370A (en) | Abnormality detection method, abnormality detection device, electronic device, and readable storage medium |
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 |