Method, device and equipment for identifying account group with abnormal resource usage

Abstract

The application relates to the computer technical field, in particular to the electronic financial technical field, and provides a resource use abnormal account group identification method, device and equipment to solve the problem that a resource use abnormal account group cannot be identified, the method comprises the following steps: obtaining a historical resource use data set corresponding to each account respectively, generating account association information based on each historical resource use data set; based on the account association information, at least one candidate account group is identified; based on each account attribute feature set corresponding to each account contained in the at least one candidate account group, a target account group is screened out from the at least one candidate account group in combination with a preset screening rule; wherein each account attribute feature set comprises self attribute features of an account and interaction attribute features of the account in an interaction process, and each target account group comprises at least one behavior abnormal account, so that the identification accuracy is improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of computers, in particular to the technical field of electronic finance, and provides a resource use abnormal account group identification method, device and equipment. BACKGROUND

[0002] In related technologies, a modularity-based community division algorithm is usually used to mine the aggregation information in a financial relationship network and determine the account groups to which the accounts in the financial relationship network belong. However, the algorithm cannot be used to screen out resource use abnormal account groups.

[0003] Therefore, the embodiments of the present application provide a new resource use abnormal account group identification method, device and equipment. SUMMARY

[0004] The embodiments of the present application provide a resource use abnormal account group identification method, device and equipment to solve the problem that resource use abnormal account groups cannot be identified.

[0005] In a first aspect, the embodiments of the present application provide a resource use abnormal account group identification method, which includes the following steps.

[0006] Obtaining a plurality of accounts from a resource use institution, and obtaining a historical resource use data set corresponding to each account, wherein each historical resource use data includes at least the interaction behavior between one account and at least one other account in a resource use process;

[0007] Performing data association processing on each historical resource use data set to generate account association information;

[0008] Performing group classification processing on the account association information to identify at least one candidate account group, wherein each candidate account group contains accounts with the same behavior in a resource use process;

[0009] Based on the account attribute feature set corresponding to each account contained in the at least one candidate account group, and in combination with a preset screening rule, a target account group is screened out from the at least one candidate account group, wherein each account attribute feature set contains the self attribute feature of one account and the interaction attribute feature of the one account in an interaction process, and each target account group contains at least one behavior abnormal account.

[0010] Optionally, the obtaining of the historical resource use data set corresponding to each account includes the following steps.

[0011] Obtaining an original resource use data set corresponding to each account;

[0012] In each of the obtained raw resource usage data sets, redundant raw resource usage data is removed, and raw resource usage data generated based on interaction behaviors between accounts and non-accounts is removed, to obtain a corresponding historical resource usage data set.

[0013] Optionally, the data association processing on the historical resource usage data set generates account association information, and includes:

[0014] Resource inflow nodes and resource outflow nodes of each historical resource usage data in each historical resource usage data set are determined respectively, each resource inflow node representing one account that transfers in resources, and each resource outflow node representing one account that transfers out resources;

[0015] Each resource inflow node is connected with a corresponding resource outflow node to form the account association information.

[0016] Optionally, the group classification processing on the account association information identifies at least one candidate account group, and includes:

[0017] For each account included in the account association information, the following operations are performed in a cyclic iteration manner until an iteration stop condition is met, and the at least one candidate account group is output:

[0018] A plurality of account cluster sets corresponding to the account at present are obtained;

[0019] For the plurality of account cluster sets, the following operations are performed respectively: obtaining a cluster evaluation value between one account cluster set in the plurality of account cluster sets and at least one associated account cluster set, wherein each associated account set is an account cluster set that has interaction behaviors with the one account cluster set in resource usage in the plurality of account cluster sets;

[0020] Based on the at least one cluster evaluation value corresponding to each of the plurality of account cluster sets, the plurality of account cluster sets are re-clustered to obtain a new plurality of account cluster sets.

[0021] Optionally, before the account association information is generated, the following operations are further included:

[0022] For each of the historical resource usage data sets, the following operations are performed respectively:

[0023] For each historical resource usage data in one historical resource usage data set, data segmentation or data analysis is performed to obtain at least one account attribute set, each account attribute set including attribute values for a same account attribute from the historical resource usage data;

[0024] Feature extraction is performed on the at least one set of account attributes to generate a set of account attribute features of the account corresponding to the one set of historical resource usage data.

[0025] Optionally, the target account group is filtered from the at least one candidate account group based on the set of account attribute features of each account included in each candidate account group and in combination with a preset filtering rule, including:

[0026] For each candidate account group in the at least one candidate account group, the following operations are respectively performed:

[0027] The set of account attribute features of each account included in each candidate account group is obtained.

[0028] For each account, the following operations are respectively performed: if more than a set number of account attribute features in the set of account attribute features of one account in the each account meet the filtering rule, the one account is determined to be an account with abnormal behavior.

[0029] The candidate account group including at least one account with abnormal behavior is taken as the target account group.

[0030] Optionally, after the target account group is determined, the following operations are further included:

[0031] The set of account attribute features of each account in the target account group is obtained.

[0032] Each set of account attribute features is input into a preset account group classification model for secondary identification to obtain an abnormal behavior probability corresponding to each account.

[0033] For each account, the following operations are respectively performed: if the abnormal behavior probability of one account in the each account exceeds a set threshold, the one account is determined to be an account with abnormal behavior.

[0034] If a total number of accounts with abnormal behavior reaches a set threshold, the target account group is determined to be a group with abnormal behavior.

[0035] In a second aspect, an embodiment of the present application further provides an account group with abnormal resource usage identification device, including:

[0036] An acquisition unit is configured to obtain a plurality of accounts from a resource usage mechanism and obtain a set of historical resource usage data corresponding to each account, wherein each set of historical resource usage data at least includes an interaction behavior between one account and at least one other account in a resource usage process.

[0037] a processing unit, configured to perform data correlation processing on each historical resource usage data set to generate account association information;

[0038] perform group classification processing on the account association information to identify at least one candidate account group, wherein each candidate account group contains accounts having the same behavior in resource usage;

[0039] a screening unit, configured to screen a target account group from the at least one candidate account group based on a set of account attribute features of each account contained in the at least one candidate account group and in combination with a preset screening rule, wherein each set of account attribute features contains self attribute features of an account and interaction attribute features of the account in an interaction process, and each target account group contains at least one account with abnormal behavior.

[0040] Optionally, the collection unit is configured to:

[0041] obtain a set of original resource usage data corresponding to each account;

[0042] remove redundant original resource usage data and original resource usage data generated based on interaction behavior between an account and a non-account from each obtained set of original resource usage data to obtain a corresponding set of historical resource usage data.

[0043] Optionally, the processing unit is configured to:

[0044] determine resource inflow nodes and resource outflow nodes of each historical resource usage data in each set of historical resource usage data, wherein each resource inflow node represents an account that transfers in resources, and each resource outflow node represents an account that transfers out resources;

[0045] connect each resource inflow node with a corresponding resource outflow node to form the account association information.

[0046] Optionally, the processing unit is configured to:

[0047] perform the following operations in a loop iteration manner for each account contained in the account association information until an iteration stop condition is met, and output the at least one candidate account group:

[0048] obtain a plurality of account clustering sets currently corresponding to the account;

[0049] For each of the plurality of account cluster sets, the following operations are performed: obtaining a cluster evaluation value between the one account cluster set and at least one associated account cluster set in the plurality of account cluster sets, wherein each associated account set is an account cluster set that has interaction behavior with the one account cluster set in the resource usage process;

[0050] Based on the obtained at least one cluster evaluation value corresponding to each of the plurality of account cluster sets, the plurality of account cluster sets are re-clustered to obtain a new plurality of account cluster sets.

[0051] Optionally, before generating the account association information, the processing unit is further configured to:

[0052] For each of the plurality of historical resource usage data sets, the following operations are performed:

[0053] For each of the historical resource usage data in one historical resource usage data set, data segmentation or data parsing is performed to obtain at least one account attribute set, and each account attribute set includes attribute values of the same account attribute from the historical resource usage data;

[0054] The at least one account attribute set is subjected to feature extraction processing to generate an account attribute feature set of an account corresponding to the one historical resource usage data set.

[0055] Optionally, the screening unit is configured to:

[0056] For each of the at least one candidate account group, the following operations are performed:

[0057] Obtain the account attribute feature set corresponding to each account included in each of the at least one candidate account group;

[0058] For each of the at least one candidate account group, the following operations are performed:

[0059] The candidate account group including at least one account with abnormal behavior is taken as the target account group.

[0060] Optionally, after determining the target account group, the screening unit is further configured to:

[0061] Obtain the account attribute feature set of each account in the target account group;

[0062] input the respective account attribute feature set into a preset account group classification model for secondary identification to obtain respective abnormal behavior probabilities of the respective accounts;

[0063] For the respective accounts, the following operations are respectively performed: if an abnormal behavior probability of one of the respective accounts exceeds a set threshold, the one account is determined to be an account with abnormal behavior;

[0064] If a total number of accounts with abnormal behavior reaches a set threshold, the target account group is determined to be a group with abnormal behavior.

[0065] In a third aspect, an embodiment of the present application further provides a computer device, including a processor and a memory, wherein the memory stores program code, when the program code is executed by the processor, the processor executes steps of any one of the above account group identification methods with abnormal resource usage.

[0066] In a fourth aspect, an embodiment of the present application further provides a computer readable storage medium, including program code, when the program product is run on a computer device, the program code is used to make the computer device execute steps of any one of the above account group identification methods with abnormal resource usage.

[0067] The present application has the following advantages:

[0068] The account group identification method, device and equipment with abnormal resource usage provided by the embodiment of the present application, the method includes: obtaining a plurality of accounts from a resource usage mechanism, obtaining respective historical resource usage data sets corresponding to the respective accounts, wherein each historical resource usage data at least includes: an interaction behavior between one account and at least one other account in a resource usage process; performing data association processing on each historical resource usage data set to generate account association information; performing group classification processing on the account association information to identify at least one candidate account group, wherein each candidate account group contains: accounts with the same behavior in the resource usage process; based on respective account attribute feature sets corresponding to the respective accounts contained in at least one candidate account group, combining a preset screening rule, screening a target account group from at least one candidate account group; wherein each account attribute feature set contains self attribute features of one account and interaction attribute features of the account in the interaction process, and each target account group contains at least one account with abnormal behavior. Compared with the related art, the account group identification method with abnormal resource usage proposed by the embodiment of the present application not only considers the topological structure information between accounts and accounts, but also introduces self attribute features of each account and interaction attribute features of each account in the interaction process, so as to identify the candidate account group with abnormal behavior from the account association information.

[0069] Other features and advantages of the present application will be set forth in the following specification, and in part will become apparent to those skilled in the art on examination of the specification or by practice of the application. The objectives and other advantages of the application will be realized and attained by the structure particularly pointed out in the written description and claims hereof as well as the appended drawings. BRIEF DESCRIPTION OF DRAWINGS

[0070] The accompanying drawings, which are included to provide a further understanding of the application and are incorporated in and constitute a part of this application, illustrate embodiments of the application and together with the description serve to explain the application. In the drawings:

[0071] Figure 1 An optional schematic diagram of one application scenario in embodiments of the present application;

[0072] Figure 2a A flowchart of identifying a resource usage abnormal account group in embodiments of the present application;

[0073] Figure 2b A schematic diagram of account association information in embodiments of the present application;

[0074] Figure 2c A flowchart of identifying a candidate account group in embodiments of the present application;

[0075] Figure 2d A schematic diagram of account association information in embodiments of the present application introducing weights of edges;

[0076] Figure 2e A schematic diagram of multiple candidate account groups in embodiments of the present application;

[0077] Figure 2f A schematic diagram of a candidate account group containing multiple money laundering accounts in embodiments of the present application;

[0078] Figure 3a A flowchart of identifying a financial fraud group in embodiments of the present application;

[0079] Figure 3b A logic diagram of identifying a financial fraud group in embodiments of the present application;

[0080] Figure 4 A structural diagram of an account group identifying device for resource usage abnormality in embodiments of the present application;

[0081] Figure 5 A structural diagram of a computer device in embodiments of the present application;

[0082] Figure 6 A structural diagram of a computing device in embodiments of the present application. DETAILED DESCRIPTION

[0083] To make the purpose, technical solutions, and advantages of the embodiments of this application more clear, the technical solutions of this application will be clearly and completely described below in conjunction with the drawings in the embodiments of this application. Obviously, the described embodiments are part of the embodiments of the technical solutions of this application, but not all of them. Based on the embodiments described in this application document, all other embodiments obtained by ordinary technicians in this field without making creative efforts shall fall within the scope of protection of the technical solutions of this application.

[0084] The terms "first," "second," and the like in the specification and claims of this application and the accompanying drawings are used to distinguish similar objects and are not necessarily used to describe a particular order or precedence. It should be understood that the terms used in this manner are interchangeable where appropriate, such that the embodiments described herein can be practiced in orders other than those illustrated or described herein.

[0085] Some of the terms used in the embodiments of the present application are explained to help those skilled in the art understand the present application.

[0086] 1. Account: An account opened by an individual user in a third-party payment institution or bank.

[0087] 2. Non-account: An account opened by an individual business operator or enterprise in a third-party payment institution or bank.

[0088] 3. Resource Use: This refers to transactions between an account and other accounts or non-accounts by transferring resources in or out. For example, an account logs into device a at 13:48 on March 21, 2021, and pays 5 yuan to the merchant to purchase one kilogram of apples. Another example is an account logs into device b at 21:58 on March 21, 2021, and receives a transfer of 10 yuan from another account. Devices a and b are both electronic devices used by users. These devices can be personal computers, mobile phones, tablets, laptops, e-book readers, smart home devices, and other computer devices with certain computing capabilities that support electronic payments.

[0089] 4. Abnormal resource usage: This refers to the frequent transfer of large amounts of resources into or out of an account within a short period of time, or the frequent transfer of large amounts of resources into and out of an account within a short period of time. For example, the illegal act of opening multiple accounts for money laundering.

[0090] 5. Community division algorithm based on modularity (Louvain algorithm):

[0091] Modularity is a commonly used algorithm to measure the quality of network community grouping. The closer the modularity is to 1, the better the quality of grouping is, that is, the better the network community division is in line with the characteristics of "close connection within the network community and relatively sparse connection outside the network community". Therefore, the optimal network community division can be obtained by maximizing the modularity.

[0092] The Louvain algorithm regards each node in the network as an independent network community, merges all connected network communities two by two, respectively calculates the modularity gain brought by each merging mode, and merges the two network communities with the maximum modularity gain into one network community. This iteration is repeated until the modularity gain no longer changes, and one or more divided network communities are obtained.

[0093] The design idea of the embodiments of the present application is briefly introduced as follows.

[0094] In the related art, the Louvain algorithm is usually used to mine the aggregation information in the financial relationship network and determine the account group to which each account in the network belongs. However, the algorithm cannot identify the account group with abnormal resource use from multiple account groups. Therefore, the embodiments of the present application provide a new account group identification method, device and equipment for identifying the account group with abnormal resource use.

[0095] The method comprises the following steps: obtaining a plurality of accounts from a resource use institution, and obtaining a historical resource use data set corresponding to each account, wherein each historical resource use data at least comprises an interaction behavior between one account and at least one other account in a resource use process; performing data association processing on each historical resource use data set to generate account association information; performing group classification processing on the account association information to identify at least one candidate account group, wherein each candidate account group comprises accounts with the same behavior in the resource use process; based on each account attribute feature set corresponding to each account included in the at least one candidate account group, and in combination with a preset screening rule, a target account group is screened from the at least one candidate account group; wherein each account attribute feature set comprises the self attribute feature of one account and the interaction attribute feature of the account in the interaction process, and each target account group comprises at least one account with abnormal behavior.

[0096] Referring to Figure 1 An application scenario schematic diagram is shown, and the application scenario comprises a first terminal device 110, a second terminal device 130 and a server 140.

[0097] The first terminal device 110 and the second terminal device 130 communicate with the server 140 through a communication network. A user logs in the application operation interface 120 through the first terminal device 110, sends a payment request to a business system of a third-party payment institution or a business system of a bank deployed on the server 140, so that the server 140 transfers digital currency of a specified amount in the payment request into a corresponding account, and completes online transaction between the user and another user. Similarly, the other user can also log in the application operation interface 120 through the second terminal device 130, send a transfer request to the server 140, so that the server 140 transfers digital currency of a specified amount in the transfer request into a corresponding account, and realizes online transaction.

[0098] In an optional embodiment, the communication network is any one of a wired network or a wireless network. Therefore, the first terminal device 110 can directly establish a communication connection with the server 140 through the wired network, or indirectly establish a communication connection with the server 140 through the wireless network, which is not limited in the present application. Similarly, the second terminal device 130 can also directly establish a communication connection with the server 140 through the wired network, or indirectly establish a communication connection with the server 140 through the wireless network, which is not limited in the present application.

[0099] Specifically, the first terminal device 110 and the second terminal device 130 in the embodiments of the present application are electronic devices used by users. The electronic devices can be personal computers, mobile phones, tablet computers, notebooks, e-book readers, smart homes, and other computer devices with certain computing power and supporting electronic payment.

[0100] The server 140 in the embodiments of the present application can be a stand-alone physical server, a server cluster or a distributed system composed of multiple physical servers, or a cloud server providing cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, content distribution networks (CDN), and basic cloud computing services such as big data and artificial intelligence platforms, which are not limited in the present application. For example, the resource usage abnormal account group identification method disclosed in the present application, wherein a plurality of servers can be composed of a blockchain, and the server is a node on the blockchain.

[0101] Referring to Figure 2a The flowchart shows that the process of identifying the resource usage abnormal account group is as follows.

[0102] S201: Obtain a plurality of accounts from a resource usage institution, and obtain a historical resource usage data set corresponding to each account, wherein each historical resource usage data at least includes an interaction behavior between one account and at least one other account in a resource usage process.

[0103] The resource usage institution refers to an institution providing a resource usage service, such as a bank, a third-party payment institution, etc. Specifically, a plurality of accounts can be obtained from the same resource usage institution, or a plurality of accounts can be obtained from different resource usage institutions, which are not limited in the present application.

[0104] Each account corresponds to a historical resource usage data set, and each historical resource usage data at least includes an interaction behavior between one account and at least one other account in a resource usage process. Optionally, the process of obtaining a historical resource usage data set corresponding to each account is as follows:

[0105] Obtain a historical resource usage data set corresponding to each account, and in each obtained historical resource usage data set, remove redundant original resource usage data and original resource usage data generated based on an interaction behavior between an account and a non-account.

[0106] An original resource usage data not only includes an interaction behavior between one account and at least one other account in a resource usage process, but also includes redundant data such as a place where the account interacts and a reason for generating the interaction behavior. For example, if an original resource usage data is "account logs in device a at 13:48 on March 21, 2021, and pays 5 yuan to a merchant for purchasing one kilogram of apples", "account logs in device a at 13:48 on March 21, 2021, and pays 5 yuan to a merchant" is taken as an interaction behavior, and "purchasing one kilogram of apples" is taken as a reason for generating the transfer behavior.

[0107] If the redundant data in the original resource usage data is retained, the total amount of data of the historical resource usage data will be increased, the time spent in processing the historical resource usage data to generate account association information will be increased, the generation accuracy of the account association information will be reduced, and further, the identification accuracy of the candidate account group will be reduced. Therefore, in the embodiments of the present application, a data cleaning method is adopted to remove the redundant data in the original resource usage data.

[0108] The resource use abnormal behavior defined in the embodiments of the present application refers to the behavior that a large amount of resources is frequently transferred into an account within a short time, or the behavior that a large amount of resources is frequently transferred out of an account within a short time, or the behavior that a large amount of resources is frequently transferred into an account and then transferred out of the account within a short time. The original resource use data generated based on the interaction behavior between the account and the non-account exactly has the characteristics of constituting the above-mentioned resource use abnormal behavior. In order to avoid misidentifying the non-account as the behavior abnormal account and affecting the division result of the final output candidate account group, the original resource use data generated based on the interaction behavior between the account and the non-account needs to be removed before step 201 is performed.

[0109] S202: Perform data association processing on each historical resource use data set to generate account association information.

[0110] Before step 202 is performed, the account attribute feature set of the corresponding account is generated based on the historical resource use data set corresponding to each account. Specifically, for each historical resource use data set, the following operations are performed respectively:

[0111] For each historical resource use data in a historical resource use data set, data segmentation or data parsing is performed to obtain at least one account attribute set, each account attribute set containing attribute values for the same account attribute from each historical resource use data; and feature extraction processing is performed on the at least one account attribute set to generate the account attribute feature set of the account corresponding to the historical resource use data set.

[0112] The traditional Louvain algorithm only considers the topological structure information between accounts, but in the resource use abnormal scenario proposed in the embodiments of the present application, the candidate account group not only presents the aggregation characteristic, but also the self attribute feature (such as the total number of devices logging into the account) of each account in the candidate account group and the interaction attribute feature (such as the total amount of monthly incoming payment of an account, the total amount of monthly outgoing payment of the account, the monthly incoming and outgoing payment ratio of the account, etc.) of each account in the interaction process can also be used as one of the bases for identifying the candidate account group. Therefore, the account attribute feature set in the embodiments of the present application at least includes the following account attribute features:

[0113] (1) the total amount of monthly incoming payment of an account;

[0114] (2) the total amount of monthly outgoing payment of the account;

[0115] (3) the number of monthly incoming payment of the account;

[0116] (4) the number of monthly outgoing payment of the account;

[0117] (5) the total number of devices logging into the account;

[0118] (6) the proportion of monthly night transactions of the account;

[0119] (7) the proportion of monthly in-out amount of the account;

[0120] (8) the proportion of monthly in-out pen number of the account;

[0121] (9) the proportion of monthly fast-in fast-out pen number of the account.

[0122] Wherein, the transactions in the time period of 00:00-6:00 and 22:00-24:00 each day are defined as night transactions; if the time interval of the transfer-in and transfer-out of the same amount of money by the account is less than 15 minutes, the transfer-in and transfer-out operation of the account for the amount of money is regarded as fast-in fast-out transaction operation.

[0123] For example, Table 1 shows the bill set of account A from February 1, 2021 to February 28, 2021, and by data segmentation or data analysis on the bill set of Table 1, the account attribute feature set of account A shown in Table 2 is obtained. The numbers in Table 1 are only illustrative and are not accurate data. The same method will be used in subsequent examples, and this content will not be repeated.

[0124] Table 1

[0125]

[0126] Table 2

[0127]

[0128] Optionally, after introducing the generation process of each account attribute feature set, the process of generating account association information is introduced as follows:

[0129] Determine the resource inflow node and the resource outflow node of each historical resource usage data in each historical resource usage data set, respectively, each resource inflow node representing an account that transfers in resources, and each resource outflow node representing an account that transfers out resources;

[0130] Connect each resource inflow node with the corresponding resource outflow node to form account association information.

[0131] For example, Table 3 shows the bill set of each of a plurality of accounts in February 2021, as well as the resource inflow node and the resource outflow node in each bill, and based on the plurality of bill sets in February 2021 shown in Table 3, the schematic diagram of the account association information shown in Table 4 is generated. Figure 2b

[0132] Table 3

[0133] ​

[0134] S203: Perform group classification processing on the account association information to identify at least one candidate account group, wherein each candidate account group includes accounts having the same behavior in resource use.

[0135] At least one candidate account group is identified from the account association information using the Louvain algorithm. Optionally, refer to Figure 2c The identification process of the candidate account group is introduced as follows with reference to the flowchart shown in the figure.

[0136] S2031: Obtain a plurality of account cluster sets currently corresponding to respective accounts.

[0137] S2032: For the plurality of account cluster sets, respectively perform the following operations: obtain a cluster evaluation value between an account cluster set in the plurality of account cluster sets and at least one associated account cluster set, wherein each associated account set is an account cluster set in the plurality of account cluster sets that has an interaction behavior with the account cluster set in resource use.

[0138] In the embodiment of the present application, the modularity gain formula as shown in formula (1) is used to obtain at least one cluster evaluation value corresponding to each of the plurality of account cluster sets.

[0139] In formula (1), m is the sum of the weights of all edges in the account association information. represents the cluster evaluation value between the account cluster set i and the associated account cluster set j; m is the sum of the weights of all edges in the account association information. represents the sum of the weights of the edges between the account cluster set i and the associated account cluster set j, represents the product of the first weight sum and the second weight sum, the first weight sum represents the sum of the weights of the edges of all associated account cluster sets connected to the account cluster set i, and the second weight sum represents the sum of the weights of the edges of all associated account cluster sets connected to the associated account cluster set j.

[0140] Formula (1);

[0141] The weight of each edge can be set to a default weight of 1, or can be flexibly set based on the transfer amount and transfer frequency between two accounts, etc. For example, if account A transfers 5 yuan to account B, the weight of the edge between account A and account B is set to 5. Since the Louvain algorithm is a mining algorithm for undirected graphs, if account A transfers 5 yuan to account B and account B transfers 10 yuan to account A, the weight of the edge between account A and account B should be set to 15 (i.e. the total of the two transfer amounts).

[0142] For example, Figure 2dThe schematic diagram of the account association information shown is an example, and the clustering evaluation value of the account clustering set A in the first round is calculated.

[0143] , , .

[0144] S2033: Based on the at least one clustering evaluation value corresponding to each of the obtained multiple account clustering sets, re-clustering the multiple account clustering sets to obtain a new multiple account clustering sets.

[0145] If the clustering evaluation value is positive, the larger the clustering evaluation value, the better the grouping quality of the new account clustering set formed by aggregating the account clustering set and the associated account clustering set generating the clustering evaluation value. Therefore, if an account clustering set corresponds to at least one positive clustering evaluation value, the account clustering set and the associated account clustering set generating the maximum clustering evaluation value are re-aggregated to form a new account clustering set.

[0146] If the clustering evaluation value is negative, it indicates that the new account clustering set formed by aggregating the account clustering set and the associated account clustering set generating the clustering evaluation value has poor grouping quality. Therefore, if the account clustering set corresponds to a negative clustering evaluation value, the account clustering set and itself are re-aggregated to form a new account clustering set.

[0147] S2034: Determine whether the multiple account clustering sets currently corresponding to all accounts no longer change, if yes, execute step 2035; otherwise, return to step 2031.

[0148] S2035: Output the new multiple account clustering sets as multiple candidate account groups.

[0149] For ease of understanding, the account association information shown in Figure 2d is obtained by using the Louvain algorithm, and the candidate account group schematic diagram shown in Figure 2e is obtained.

[0150] Specifically, (1) each account is taken as an account clustering set in the first round, and the clustering evaluation value between each account clustering set and at least one associated account clustering set is calculated.

[0151] , , , , , , .

[0152] After the first round of calculation, for the account cluster set to which account A belongs, the maximum value of the cluster evaluation is evaluated, then the account cluster set to which account A belongs and the account cluster set to which account B belongs are re-clustered to obtain a new account cluster set; for the account cluster set to which account B belongs, the maximum value of the cluster evaluation is evaluated, then the account cluster set to which account B belongs and the account cluster set to which account A belongs are re-clustered to obtain a new account cluster set. For the account cluster sets to which other accounts belong, the above steps are also performed, then after the first round, the output new account cluster sets are three, which are (A, B), (C, D) and (E, F) respectively.

[0153] (2) Take (A, B), (C, D) and (E, F) as the second round of account cluster sets, and calculate the cluster evaluation value between each account cluster set and at least one associated account cluster set.

[0154] , , .

[0155] After the second round of calculation, the cluster evaluation value of each account cluster set is a negative number, and the accounts contained in each account cluster set remain unchanged, and finally (A, B), (C, D) and (E, F) are output as candidate account groups.

[0156] S204: Based on each account attribute feature set corresponding to each account contained in at least one candidate account group, and in combination with a preset screening rule, a target account group is screened from the at least one candidate account group; wherein each account attribute feature set contains the self attribute feature of an account and the interaction attribute feature of the account in the interaction process, and each target account group contains at least one behaviorally abnormal account.

[0157] Optionally, for each candidate account group in the at least one candidate account group, the following operations are respectively performed:

[0158] Obtain each account attribute feature set corresponding to each account contained in a candidate account group in each candidate account group;

[0159] For each account, the following operations are respectively performed: if more than a set number of account attribute features in the account attribute feature set of one account in each account meet the screening rule, the account is determined to be a behaviorally abnormal account;

[0160] The candidate account group containing at least one behaviorally abnormal account is taken as the target account group.

[0161] The abnormal resource usage behavior specified in the embodiments of the present application refers to the behavior of an account frequently transferring a large amount of resources in a short period of time, or the behavior of an account frequently transferring a large amount of resources out in a short period of time, or the behavior of an account frequently transferring a large amount of resources in and then transferring a large amount of resources out in a short period of time.

[0162] The screening rules in the embodiment of the present application include at least the following rules:

[0163] (1) The total monthly deposit amount of an account is greater than the total monthly deposit amount threshold;

[0164] (2) The total monthly withdrawal amount of the account is greater than the total monthly withdrawal amount threshold;

[0165] (3) The monthly deposit number of the account exceeds the monthly deposit number threshold;

[0166] (4) The monthly withdrawal number of the account exceeds the monthly withdrawal threshold;

[0167] (5) The total number of devices logged into the account exceeds the total device threshold;

[0168] (6) The monthly nighttime trading ratio of the account is greater than the monthly nighttime trading ratio threshold;

[0169] (7) The monthly deposit and withdrawal ratio of the account is greater than the monthly deposit and withdrawal ratio;

[0170] (8) The monthly deposit and withdrawal ratio of the account is greater than the monthly deposit and withdrawal ratio;

[0171] (9) The monthly fast-in-fast-out ratio of the account is greater than the monthly fast-in-fast-out ratio threshold.

[0172] Therefore, if more than a set number of account attribute features in the account attribute feature set of an account meet the screening rules, the account is determined to be an account with abnormal behavior.

[0173] For example, in an anti-money laundering scenario, if more than 80% of an account's attribute characteristics meet the screening rules, the account is determined to be a money laundering account, and the candidate account group to which the account belongs may be determined to be a suspected money laundering group.

[0174] For example, in the anti-money laundering scenario, Figure 2f The candidate account group shown contains four accounts, three of which are determined to be money laundering accounts. All three money laundering accounts have transferred multiple large sums of money to the unclassified accounts in the group. Therefore, based on the interaction between the unclassified accounts and other accounts in the group, the classified accounts can be determined to be money laundering accounts.

[0175] Since the target account group of the resource use abnormal scenario has the aggregation characteristic, a candidate account group including at least one account with behavior abnormality is taken as the target account group. Compared with the related art, the account group identification method for resource use abnormality provided in the embodiment of the application not only considers the topological structure information between accounts, but also introduces the self attribute feature of each account and the interaction attribute feature of each account in the interaction process, so as to accurately identify the candidate account group with behavior abnormal accounts from the account association information.

[0176] Since the granularity of the first screening is coarse, multiple target account groups can be screened out. In order to further improve the identification accuracy, the embodiment of the application further provides secondary screening with finer granularity. Specifically, after the target account group is determined, the account attribute feature set of each account in the target account group can be obtained, and each account attribute feature set is input into a preset account group classification model for secondary identification to obtain the respective abnormal behavior probability of each account.

[0177] For each account, the following operations are respectively performed: if the abnormal behavior probability of one account in each account exceeds a set threshold, the account is determined as a behavior abnormal account.

[0178] If the total number of behavior abnormal accounts reaches a set threshold, the target account group is determined as a behavior abnormal group.

[0179] For example, in the anti-money laundering scenario, assuming that the suspected money laundering group contains 10 accounts, the respective account attribute feature set of the 10 accounts is obtained, each account attribute feature set is input into the account group classification model for secondary identification, if 8 accounts of the 10 accounts are identified as money laundering accounts, the suspected money laundering group is determined as a money laundering group; if 2 accounts of the 10 accounts are identified as money laundering accounts, the suspected money laundering group is determined as a normal group.

[0180] The account group classification model is a binary classification model, a large amount of sample data is used to train the account group classification model, in the stage when the model is not trained, the secondary screening is performed in the manner of manual review, and when the model is trained, the secondary screening is performed using the model. In the embodiment of the application, the account group classification model can realize the function of automatically identifying the behavior abnormal group end to end, simplifies the identification process of the behavior abnormal group, saves the identification time, improves the work efficiency and the identification accuracy.

[0181] Wherein, during the model cold start period, the data of the account group audited by the artificial is taken as the negative sample data, and the data of the account group not audited by the artificial is taken as the positive sample data; after the model training is completed, the data of the group determined as the behavior abnormal by the model is taken as the negative sample data, and the data of the group determined as the normal by the model is taken as the positive sample data; every time interval, the parameters in the model are retrained.

[0182] Whether the sample data is positive or negative, it contains the following two parts: the first part is the account attribute feature set corresponding to each account in the account group, and the second part is whether the account group is marked as the target account group in the first round of screening.

[0183] The embodiments of the present application can also be applied to the anti-financial fraud scene, refer to Figure 3a The flowchart and Figure 3b The logical diagram is shown, and the process of identifying the financial fraud group is introduced.

[0184] S301: Obtain the 2020 third quarter bill set corresponding to each of the 10 accounts;

[0185] S302: Based on each 2020 third quarter bill set, generate account association information;

[0186] S303: Using Louvain algorithm, determine the account group to which each account in the account association information belongs, and obtain 4 candidate account groups;

[0187] S304: Based on the account attribute feature set corresponding to each account in each candidate account group, combined with the screening rule, 1 candidate account group is identified as a normal group, and the other 3 candidate account groups are identified as suspected financial fraud groups;

[0188] S305: Using the trained account group classification model, the three suspected financial fraud groups are subjected to secondary screening, the first two suspected financial fraud groups are identified as normal groups, and the last suspected financial fraud group is identified as a financial fraud group.

[0189] Refer to Figure 4 The structure diagram is shown, and the account group identification device 400 of the resource use abnormality includes a collection unit 401, a processing unit 402 and a screening unit 403, wherein,

[0190] The collection unit 401 is used for obtaining a plurality of accounts from a resource use institution, and obtaining a historical resource use data set corresponding to each account, wherein each historical resource use data at least includes: the interaction behavior between an account and at least one other account in the resource use process;

[0191] The processing unit 402 is configured to perform data correlation processing on each historical resource usage data set to generate account correlation information.

[0192] The account correlation information is subjected to group classification processing to identify at least one candidate account group, wherein each candidate account group contains accounts having the same behavior in resource usage.

[0193] The screening unit 403 is configured to screen a target account group from the at least one candidate account group based on a set of account attribute features of each account included in the at least one candidate account group and a preset screening rule, wherein each set of account attribute features contains self attribute features of an account and interaction attribute features of the account in an interaction process, and each target account group contains at least one account with abnormal behavior.

[0194] Optionally, the collection unit 401 is configured to:

[0195] obtain a set of original resource usage data corresponding to each account;

[0196] remove redundant original resource usage data and original resource usage data generated based on interaction between an account and a non-account from each obtained set of original resource usage data to obtain a corresponding set of historical resource usage data.

[0197] Optionally, the processing unit 402 is configured to:

[0198] determine resource inflow nodes and resource outflow nodes of each historical resource usage data in each set of historical resource usage data, wherein each resource inflow node represents an account that transfers in resources, and each resource outflow node represents an account that transfers out resources;

[0199] connect each resource inflow node with a corresponding resource outflow node to form the account correlation information.

[0200] Optionally, the processing unit 402 is configured to:

[0201] for each account included in the account correlation information, perform the following operations in a loop iteration manner until an iteration stop condition is met, and output the at least one candidate account group:

[0202] obtain a plurality of account clustering sets currently corresponding to the account;

[0203] For each of the plurality of account cluster sets, the following operations are performed: obtaining a cluster evaluation value between the one account cluster set and at least one associated account cluster set, wherein each associated account set is an account cluster set that has interaction behavior with the one account cluster set in resource usage in the plurality of account cluster sets;

[0204] Based on the obtained at least one cluster evaluation value corresponding to each of the plurality of account cluster sets, the plurality of account cluster sets are re-clustered to obtain a new plurality of account cluster sets.

[0205] Optionally, before generating the account association information, the processing unit 402 is further configured to:

[0206] For each of the plurality of historical resource usage data sets, the following operations are performed:

[0207] For each of the historical resource usage data in one historical resource usage data set, data segmentation or data parsing is performed to obtain at least one account attribute set, and each account attribute set includes attribute values of the same account attribute from the historical resource usage data;

[0208] The at least one account attribute set is subjected to feature extraction processing to generate an account attribute feature set of an account corresponding to the one historical resource usage data set.

[0209] Optionally, the screening unit 403 is configured to:

[0210] For each of the at least one candidate account group, the following operations are performed:

[0211] Obtaining the account attribute feature set corresponding to each account included in the one candidate account group in the each candidate account group;

[0212] For each of the accounts, the following operations are performed: if more than a set number of account attribute features in the account attribute feature set of one account in the accounts meet the screening rule, the one account is determined to be an abnormal account;

[0213] The candidate account group including at least one abnormal account is taken as the target account group.

[0214] Optionally, after determining the target account group, the screening unit 403 is further configured to:

[0215] Obtaining the account attribute feature set of each account in the target account group;

[0216] input the respective account attribute feature sets into a preset account group classification model for secondary identification, to obtain respective abnormal behavior probabilities of the respective accounts;

[0217] For the respective accounts, the following operations are respectively performed: if an abnormal behavior probability of one of the respective accounts exceeds a set threshold, the one account is determined to be an account with abnormal behavior;

[0218] If a total number of accounts with abnormal behavior reaches a set threshold, the target account group is determined to be a group with abnormal behavior.

[0219] For the sake of description, each part is described as a module (or unit) according to function. Of course, the functions of the modules (or units) can be implemented in the same or multiple software or hardware in the implementation of the present application.

[0220] After introducing the resource usage abnormal account group identification method and device of the example embodiment of the present application, next, the computer device according to another example embodiment of the present application is introduced.

[0221] Those skilled in the art can understand that each aspect of the present application can be implemented as a system, a method or a program product. Therefore, each aspect of the present application can be specifically implemented as a complete hardware embodiment, a complete software embodiment (including firmware, microcode, etc.), or an embodiment combined with hardware and software, which can be collectively referred to as "circuit", "module" or "system".

[0222] Based on the same inventive concept as the above method embodiment, the present embodiment also provides a computer device, which is described with reference to Figure 5 As shown in the figure, the computer device 500 can at least include a processor 501 and a memory 502. The memory 502 stores program code, and when the program code is executed by the processor 501, the processor 501 executes the steps of any one of the above resource usage abnormal account group identification methods.

[0223] In some possible embodiments, the computing device according to the present application can at least include at least one processor and at least one memory. The memory stores program code, and when the program code is executed by the processor, the processor executes the steps of the resource usage abnormal account group identification method according to various example embodiments of the present application described above in the specification. For example, the processor can execute the steps as shown in Figure 2a

[0224] The computing device 600 according to this embodiment of the present application is described below with reference to Figure 6 Figure 6 ​​The computing device 600 is merely an example and should not limit the functionality and scope of use of the embodiments of the present application.

[0225] like Figure 6 As shown, computing device 500 is implemented as a general-purpose computing device. Components of computing device 600 may include, but are not limited to, at least one processing unit 601, at least one storage unit 602, and a bus 603 connecting various system components (including storage unit 602 and processing unit 601).

[0226] Bus 603 represents one or more of several types of bus structures, including a memory bus or memory controller, a peripheral bus, and a processor or local bus using any of a variety of bus architectures.

[0227] The storage unit 602 may include a readable medium in the form of a volatile memory, such as a random access memory (RAM) 6021 and / or a cache memory unit 6022 , and may further include a read-only memory (ROM) 6023 .

[0228] The storage unit 602 may also include a program / utility 6025 having a set (at least one) of program modules 6024, such program modules 6024 including but not limited to: an operating system, one or more application programs, other program modules and program data, each of which or some combination may include an implementation of a network environment.

[0229] Computing device 600 may also communicate with one or more external devices 604 (e.g., a keyboard, pointing device, etc.), one or more devices that enable a user to interact with computing device 600, and / or any device that enables computing device 600 to communicate with one or more other computing devices (e.g., a router, modem, etc.). This communication may occur via input / output (I / O) interface 605. Furthermore, computing device 600 may communicate with one or more networks (e.g., a local area network (LAN), a wide area network (WAN), and / or a public network such as the Internet) via network adapter 606. As shown, network adapter 606 communicates with other modules of computing device 600 via bus 603. It should be understood that, although not shown, other hardware and / or software modules may be used in conjunction with computing device 600, including but not limited to microcode, device drivers, redundant processors, external disk drive arrays, RAID systems, tape drives, and data backup storage systems.

[0230] As the method embodiments described above are based on the same inventive concept, the various aspects of the account group identification method for resource usage anomaly provided by the present application can also be implemented in the form of a program product, which includes program codes. When the program product is run on a computer device, the program codes are used to cause the computer device to perform the steps of the account group identification method for resource usage anomaly according to various exemplary embodiments of the present application described above in the specification, for example, the electronic device can perform the steps as shown in Figure 2a FIG. 13.

[0231] The program product can adopt any combination of one or more readable media. The readable medium can be a readable signal medium or a readable storage medium. The readable storage medium may, for example, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or apparatus, or any combination of the above. More specific examples (a non-exhaustive list) of the readable storage medium include an electrical connection having one or more wires, a portable disc, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above.

[0232] Although the preferred embodiments of the present application have been described, those skilled in the art can make additional changes and modifications to the embodiments once they have the basic inventive concept. Therefore, the appended claims are intended to be interpreted as including all the preferred embodiments and all changes and modifications falling within the scope of the present application.

[0233] Obviously, those skilled in the art can make various modifications and variations to the present application without departing from the spirit and scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the claims of the present application and their equivalents, the present application is also intended to include these modifications and variations.

Claims

1. A method for identifying account groups with abnormal resource usage, characterized in that: include: Acquire multiple accounts from a resource-using organization and obtain a set of historical resource usage data corresponding to each account, wherein each set of historical resource usage data includes at least: interaction behavior between one account and at least one other account during resource usage; Perform data association processing on each historical resource usage data set to generate account association information; For each account included in the account association information, the following operations are performed in a loop iteration manner until an iteration stop condition is satisfied, and at least one candidate account group is output, wherein each candidate account group includes: accounts having the same behavior during resource usage; obtaining multiple account cluster sets currently corresponding to each of the accounts; for each of the multiple account cluster sets, the following operations are performed: obtaining a cluster evaluation value between one of the multiple account cluster sets and at least one associated account cluster set, wherein each associated account set is: an account cluster set in the multiple account cluster sets that has an interactive behavior with the one account cluster set during resource usage; based on the obtained at least one cluster evaluation value corresponding to each of the multiple account cluster sets, the multiple account cluster sets are re-clustered to obtain multiple new account cluster sets; Based on the account attribute feature sets corresponding to each account included in the at least one candidate account group, combined with preset screening rules, a target account group is screened out from the at least one candidate account group; wherein each account attribute feature set includes the inherent attribute features of an account and the interaction attribute features of the account during the interaction process, and each target account group includes at least one account with abnormal behavior.

2. The method according to claim 1, wherein The step of obtaining a set of historical resource usage data corresponding to each account includes: Obtaining a set of original resource usage data corresponding to each of the accounts; From each of the original resource usage data sets obtained, redundant original resource usage data and original resource usage data generated based on interaction between accounts and non-accounts are eliminated to obtain a corresponding historical resource usage data set.

3. The method according to claim 1, wherein The performing data association processing on the historical resource usage data set to generate account association information includes: Determine the resource inflow node and resource outflow node of each historical resource usage data in each historical resource usage data set, each resource inflow node represents an account for transferring in resources, and each resource outflow node represents an account for transferring out resources; Each resource inflow node is connected to the corresponding resource outflow node to form the account association information.

4. The method according to claim 1, wherein Before generating account association information, it also includes: For each of the historical resource usage data sets, perform the following operations: Performing data segmentation or data parsing on each historical resource usage data in a historical resource usage data set to obtain at least one account attribute set, each account attribute set including: attribute values ​​for the same account attribute from each historical resource usage data; Perform feature extraction processing on the at least one account attribute set to generate an account attribute feature set of the account corresponding to the one historical resource usage data set.

5. The method according to claim 1, wherein The step of screening out a target account group from the at least one candidate account group based on the account attribute feature set corresponding to each account included in the at least one candidate account group and in combination with a preset screening rule includes: For each candidate account group in the at least one candidate account group, perform the following operations: Obtaining, in each candidate account group, a set of account attribute features corresponding to each account included in a candidate account group; For each of the accounts, the following operations are performed: if more than a set number of account attribute features in the account attribute feature set of one of the accounts meet the screening rule, the account is determined to be an account with abnormal behavior; A candidate account group including at least one account with abnormal behavior is used as the target account group.

6. The method according to any one of claims 1 to 5, wherein: After determining the target account group, the method further includes: Obtaining a set of account attribute characteristics for each account in the target account group; Input each account attribute feature set into a preset account group classification model for secondary identification to obtain the abnormal behavior probability corresponding to each account; For each of the accounts, the following operations are performed: if the abnormal behavior probability of one of the accounts exceeds a set threshold, the account is determined to be an account with abnormal behavior; If the total number of accounts with abnormal behavior reaches a set threshold, the target account group is determined to be a group with abnormal behavior.

7. A device for identifying account groups with abnormal resource usage, characterized in that: include: A collection unit is configured to obtain a plurality of accounts from a resource-using organization and obtain a set of historical resource usage data corresponding to each account, wherein each set of historical resource usage data includes at least: interaction behavior between one account and at least one other account during resource usage; A processing unit, configured to perform data association processing on each historical resource usage data set to generate account association information; For each account included in the account association information, the following operations are performed in a loop iteration manner until an iteration stop condition is satisfied, and at least one candidate account group is output, wherein each candidate account group includes: accounts having the same behavior during resource usage; obtaining multiple account cluster sets currently corresponding to each of the accounts; for each of the multiple account cluster sets, the following operations are performed: obtaining a cluster evaluation value between one of the multiple account cluster sets and at least one associated account cluster set, wherein each associated account set is: an account cluster set in the multiple account cluster sets that has an interactive behavior with the one account cluster set during resource usage; based on the obtained at least one cluster evaluation value corresponding to each of the multiple account cluster sets, the multiple account cluster sets are re-clustered to obtain multiple new account cluster sets; The screening unit is used to screen out a target account group from the at least one candidate account group based on the account attribute feature set corresponding to each account included in the at least one candidate account group, in combination with preset screening rules; wherein each account attribute feature set includes the inherent attribute features of an account and the interaction attribute features of the account during the interaction process, and each target account group includes at least one account with abnormal behavior.

8. The device according to claim 7, wherein The acquisition unit is used for: Obtaining a set of original resource usage data corresponding to each of the accounts; From each of the original resource usage data sets obtained, redundant original resource usage data and original resource usage data generated based on interaction between accounts and non-accounts are eliminated to obtain a corresponding historical resource usage data set.

9. The device according to claim 7, wherein The processing unit is used for: Determine the resource inflow node and resource outflow node of each historical resource usage data in each historical resource usage data set, each resource inflow node represents an account for transferring in resources, and each resource outflow node represents an account for transferring out resources; Each resource inflow node is connected to the corresponding resource outflow node to form the account association information.

10. The device according to claim 7, wherein Before generating the account association information, the processing unit is further configured to: For each of the historical resource usage data sets, perform the following operations: Performing data segmentation or data parsing on each historical resource usage data in a historical resource usage data set to obtain at least one account attribute set, each account attribute set including: attribute values ​​for the same account attribute from each historical resource usage data; Perform feature extraction processing on the at least one account attribute set to generate an account attribute feature set of the account corresponding to the one historical resource usage data set.

11. The device according to claim 7, wherein The screening unit is used to: For each candidate account group in the at least one candidate account group, perform the following operations: Obtaining, in each candidate account group, a set of account attribute features corresponding to each account included in a candidate account group; For each of the accounts, the following operations are performed: if more than a set number of account attribute features in the account attribute feature set of one of the accounts meet the screening rule, the account is determined to be an account with abnormal behavior; A candidate account group including at least one account with abnormal behavior is used as the target account group.

12. The device according to any one of claims 7 to 11, characterized in that After determining the target account group, the screening unit is further configured to: Obtaining a set of account attribute characteristics for each account in the target account group; Input each account attribute feature set into a preset account group classification model for secondary identification to obtain the abnormal behavior probability corresponding to each account; For each of the accounts, the following operations are performed: if the abnormal behavior probability of one of the accounts exceeds a set threshold, the account is determined to be an account with abnormal behavior; If the total number of accounts with abnormal behavior reaches a set threshold, the target account group is determined to be a group with abnormal behavior.

13. A computer device, characterized in that: The method comprises a processor and a memory, wherein the memory stores program codes, and when the program codes are executed by the processor, the processor executes the steps of the method according to any one of claims 1 to 6.

14. A computer-readable storage medium, characterized in that The method comprises a program code, and when the program product is run on a computer device, the program code is used to enable the computer device to execute the steps of the method according to any one of claims 1 to 6.

Images