Information processing method and device

Abstract

The present invention discloses a method and device for information processing, which includes obtaining transaction data of all cardholders within a preset period, analyzing the transaction data of all cardholders within the preset period, determining multiple cardholder relationship network communities, determining credit card consumption transaction data and debit card payment transaction data in the cardholder relationship network community for each cardholder relationship network community, and determining abnormal data in the cardholder relationship network community based on the credit card consumption transaction data and the debit card payment transaction data. By obtaining the transaction data of all cardholders of each commercial bank within the preset period, dividing each transaction data into communities based on graph calculation, obtaining transaction data belonging to the same community, and then performing abnormal data detection on the transaction data of the same community, the accuracy and coverage of abnormal data detection can be improved, and the efficiency of abnormal data detection of the system can be improved, and the power consumption of the system can be reduced.

Description

Technical Field

[0001] The embodiments of the present invention relate to the field of data processing technology, and in particular, to a method and device for information processing. Background Art

[0002] Abnormal use of credit cards can enable cardholders to obtain high amounts of funds in a short period of time. Under the temptation of high returns, abnormal use behavior has been repeatedly banned, increasing the risks of issuing banks and increasing the instability of my country's financial order. How to accurately identify abnormal data from the cardholder's transaction data, deeply explore the customer's consumption repayment ability and credit level, better provide card issuance services, and obtain more profit points; as well as crack down on fake merchants and institutions with large-scale abnormal use and purify the payment environment, has become a new focus of major banks and UnionPay. However, the data barriers between commercial banks are high, and each is an isolated information island. They can only identify abnormal use behavior based on the characteristics of their own transaction data. There is a problem that they cannot track the whereabouts of funds in cross-bank transactions and cannot cover abnormal use scenarios of multiple cards for one person or multiple cards for multiple people. Summary of the invention

[0003] The embodiments of the present invention provide a method and device for information processing, so as to improve the accuracy and coverage of information processing.

[0004] In a first aspect, an embodiment of the present invention provides an information processing method, including:

[0005] Obtain transaction data of all cardholders within a preset period;

[0006] Analyzing the transaction data of all cardholders within the preset time period to determine a plurality of cardholder relationship network communities;

[0007] For each cardholder relationship network community, the credit card consumption transaction data and the debit card agency payment transaction data in the cardholder relationship network community are determined; based on the credit card consumption transaction data and the debit card agency payment transaction data, the abnormal data in the cardholder relationship network community is determined.

[0008] In the above technical solution, by obtaining the transaction data of all cardholders of each commercial bank within a preset period of time, and dividing the transaction data into communities based on graph calculation, the transaction data belonging to the same community is obtained, and then abnormal data detection is performed on the transaction data of the same community. This can improve the accuracy and coverage of abnormal data detection, and at the same time improve the efficiency of system abnormal data detection and reduce system power consumption.

[0009] Optionally, analyzing the transaction data of all cardholders within the preset period to determine a plurality of cardholder relationship network communities includes:

[0010] Obtaining transaction data of all cardholders within the preset time period; the transaction data includes bank card number, mobile phone number, ID card number, device number and application user number;

[0011] The bank card number, mobile phone number, ID number, device number and application user number in the transaction data are used as vertices, and the mobile phone number, ID number, device number and application user number in the same transaction data are connected to the bank card number to obtain a cardholder relationship network diagram;

[0012] The cardholder relationship network diagram is divided into communities to obtain a plurality of cardholder relationship network communities.

[0013] Optionally, dividing the cardholder relationship network graph into communities to obtain a plurality of cardholder relationship network communities includes:

[0014] According to the cardholder relationship network diagram, traverse each vertex, determine multiple cardholder relationship network diagrams with connecting edges and set the community initial attribute value of each vertex;

[0015] Iteratively updating the community initial attribute value of each vertex in each cardholder relationship network graph with connected edges until the community attribute values ​​of all vertices in each cardholder relationship network graph with connected edges are changed to the minimum community attribute value in each cardholder relationship network graph with connected edges, and determining the minimum community attribute value as the community ID of each cardholder relationship network graph with connected edges; wherein each iterative update is to update the community attribute value of the vertex to the minimum value of the community attribute values ​​of all first-order neighbor vertices and its own community attribute value;

[0016] The community ID is associated with the transaction data of each vertex in the cardholder relationship network graph having a connecting edge, so as to obtain the plurality of cardholder relationship network communities.

[0017] Optionally, the iterative updating of the community initial attribute value of each vertex in each cardholder relationship network graph having a connecting edge includes:

[0018] Count the number of first-order neighbor vertices of each vertex;

[0019] If the number of the first-order neighbor vertices is less than a preset data processing threshold, the first-order neighbor vertices of each vertex are recorded in the first set to iteratively update the initial attribute value of the community; otherwise, the first-order neighbor vertices of each vertex are recorded in the second set and initialized to the driver; and the vertices in the second set are broadcast to other thread pools (executors), so that the other executors obtain vertices from the second set to iteratively update the initial attribute value of the community.

[0020] Optionally, determining abnormal data in the cardholder relationship network community based on the credit card consumption transaction data and the debit card payment transaction data includes:

[0021] Determining, from the credit card consumption transaction data and debit card agency payment transaction data in the cardholder relationship network community, the credit card consumption transaction data and debit card agency payment transaction data whose percentage of amount difference is less than a first threshold and whose absolute value of transaction time difference is less than a second threshold, and determining the credit card consumption transaction data and debit card agency payment transaction data whose percentage of amount difference is less than the first threshold and whose absolute value of transaction time difference is less than the second threshold as abnormal data in the cardholder relationship network community;

[0022] After removing the credit card consumption transaction data and debit card agency payment transaction data whose percentage of the amount difference is less than the first threshold and whose absolute value of the transaction time difference is less than the second threshold from the credit card consumption transaction data and debit card agency payment transaction data in the cardholder relationship network community, traverse each agency payment transaction data to obtain its corresponding community ID, agency payment amount and agency payment time;

[0023] Traverse the consumption transaction data of each credit card under the community ID to obtain a transaction set;

[0024] Determine from the transaction set a transaction set subset having the smallest absolute value of the difference between the sum of the transaction amounts and the payment amount;

[0025] If the percentage of the absolute value of the difference between the sum of the transaction amounts of the transaction set subset and the payment amount is less than the first threshold and the difference between the transaction time and the payment time in the transaction set subset is less than the second threshold, the consumption transaction data of the credit card in the transaction set subset and the traversed payment transaction data are determined as abnormal data in the cardholder relationship network community.

[0026] In a second aspect, an embodiment of the present invention provides an information processing device, including:

[0027] An acquisition unit, used to acquire transaction data of all cardholders within a preset period of time;

[0028] The processing unit is used to analyze the transaction data of all cardholders within the preset time period to determine a plurality of cardholder relationship network communities; for each cardholder relationship network community, determine the credit card consumption transaction data and the debit card payment transaction data in the cardholder relationship network community; and determine the abnormal data in the cardholder relationship network community based on the credit card consumption transaction data and the debit card payment transaction data.

[0029] Optionally, the processing unit is specifically used for:

[0030] Obtaining transaction data of all cardholders within the preset time period; the transaction data includes bank card number, mobile phone number, ID card number, device number and application user number;

[0031] The bank card number, mobile phone number, ID number, device number and application user number in the transaction data are used as vertices, and the mobile phone number, ID number, device number and application user number in the same transaction data are connected to the bank card number to obtain a cardholder relationship network diagram;

[0032] The cardholder relationship network diagram is divided into communities to obtain a plurality of cardholder relationship network communities.

[0033] Optionally, the processing unit is specifically used for:

[0034] According to the cardholder relationship network diagram, traverse each vertex, determine multiple cardholder relationship network diagrams with connecting edges and set the community initial attribute value of each vertex;

[0035] Iteratively updating the community initial attribute value of each vertex in each cardholder relationship network graph with connected edges until the community attribute values ​​of all vertices in each cardholder relationship network graph with connected edges are changed to the minimum community attribute value in each cardholder relationship network graph with connected edges, and determining the minimum community attribute value as the community ID of each cardholder relationship network graph with connected edges; wherein each iterative update is to update the community attribute value of the vertex to the minimum value of the community attribute values ​​of all first-order neighbor vertices and its own community attribute value;

[0036] The community ID is associated with the transaction data of each vertex in the cardholder relationship network graph having a connecting edge, so as to obtain the plurality of cardholder relationship network communities.

[0037] Optionally, the processing unit is specifically used for:

[0038] Count the number of first-order neighbor vertices of each vertex;

[0039] If the number of the first-order neighbor vertices is less than a preset data processing threshold, the first-order neighbor vertices of each vertex are recorded in the first set to iteratively update the initial attribute value of the community; otherwise, the first-order neighbor vertices of each vertex are recorded in the second set and initialized to the driver; and the vertices in the second set are broadcast to other executors, so that the other executors obtain vertices from the second set to iteratively update the initial attribute value of the community.

[0040] Optionally, the processing unit is specifically used for:

[0041] Determining, from the credit card consumption transaction data and debit card agency payment transaction data in the cardholder relationship network community, the credit card consumption transaction data and debit card agency payment transaction data whose percentage of amount difference is less than a first threshold and whose absolute value of transaction time difference is less than a second threshold, and determining the credit card consumption transaction data and debit card agency payment transaction data whose percentage of amount difference is less than the first threshold and whose absolute value of transaction time difference is less than the second threshold as abnormal data in the cardholder relationship network community;

[0042] After removing the credit card consumption transaction data and debit card agency payment transaction data whose percentage of the amount difference is less than the first threshold and whose absolute value of the transaction time difference is less than the second threshold from the credit card consumption transaction data and debit card agency payment transaction data in the cardholder relationship network community, traverse each agency payment transaction data to obtain its corresponding community ID, agency payment amount and agency payment time;

[0043] Traverse the consumption transaction data of each credit card under the community ID to obtain a transaction set;

[0044] Determine from the transaction set a transaction set subset having the smallest absolute value of the difference between the sum of the transaction amounts and the payment amount;

[0045] If the percentage of the absolute value of the difference between the sum of the transaction amounts of the transaction set subset and the payment amount is less than the first threshold and the difference between the transaction time and the payment time in the transaction set subset is less than the second threshold, the consumption transaction data of the credit card in the transaction set subset and the traversed payment transaction data are determined as abnormal data in the cardholder relationship network community.

[0046] In a third aspect, an embodiment of the present invention further provides a computing device, including:

[0047] A memory for storing program instructions;

[0048] The processor is used to call the program instructions stored in the memory and execute the above information processing method according to the obtained program.

[0049] In a fourth aspect, an embodiment of the present invention further provides a computer-readable non-volatile storage medium, comprising computer-readable instructions, which enables the computer to perform the above-mentioned information processing method when the computer reads and executes the computer-readable instructions. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0051] Figure 1 A schematic diagram of a system architecture provided for an embodiment of the present invention;

[0052] Figure 2 A flowchart of an information processing method provided by an embodiment of the present invention;

[0053] Figure 3 A schematic diagram of a cardholder relationship network provided by an embodiment of the present invention;

[0054] Figure 4 A schematic diagram of iterative update of community attribute values ​​provided by an embodiment of the present invention;

[0055] Figure 5 A schematic diagram of iterative update of community attribute values ​​provided by an embodiment of the present invention;

[0056] Figure 6 A schematic diagram of the structure of an information processing device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0057] In order to make the purpose, technical scheme and advantages of the present invention clearer, the present invention will be further described in detail below with reference to the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0058] Figure 1 A system architecture applicable to an embodiment of the present invention is exemplarily shown. The system architecture may be a server 100 . The server 100 may include a processor 110 , a communication interface 120 , and a memory 130 .

[0059] The communication interface 120 is used to communicate with the servers of major commercial banks, and to send and receive data transmitted by the servers of major commercial banks to achieve communication.

[0060] The processor 110 is the control center of the server 100, and uses various interfaces and routes to connect various parts of the entire server 100, and executes various functions of the server 100 and processes data by running or executing software programs and / or modules stored in the memory 130, and calling data stored in the memory 130. Optionally, the processor 110 may include one or more processing units.

[0061] The memory 130 can be used to store software programs and modules. The processor 110 executes various functional applications and data processing by running the software programs and modules stored in the memory 130. The memory 130 can mainly include a program storage area and a data storage area, wherein the program storage area can store an operating system, an application required for at least one function, etc.; the data storage area can store data created according to business processing, etc. In addition, the memory 130 can include a high-speed random access memory, and can also include a non-volatile memory, such as at least one disk storage device, a flash memory device, or other volatile solid-state storage devices.

[0062] It should be noted that the above Figure 1 The structure shown is only an example and is not limited to this embodiment of the present invention.

[0063] Based on the above description, Figure 2 The flow of an information processing method provided by an embodiment of the present invention is shown in detail. The flow can be executed by an information processing device, which can be located in the above-mentioned server 100 or can also be the server 100.

[0064] like Figure 2 As shown in the figure, the process specifically includes:

[0065] Step 201, obtaining transaction data of all cardholders within a preset time period.

[0066] In an embodiment of the present invention, transaction data of all cardholders of various commercial banks can be obtained, wherein the transaction data may include bank card number, mobile phone number, ID number, device number and application user number, wherein the application user number may be a UnionPay user number.

[0067] China UnionPay is a bank card association in China. Through the UnionPay interbank transaction clearing system, it realizes the interconnection and resource sharing between commercial bank systems, ensuring the use of bank cards across banks, regions and borders. China UnionPay has a large amount of cardholder transaction consumption data, especially a large amount of interbank transaction data. Such a huge amount of data can break the current situation of information islands between commercial banks.

[0068] Through various communication interfaces, the cardholder-related information in UnionPay's major transaction systems, including omni-channel card binding, omni-channel merchant card binding, Cloud QuickPass APP registration and card binding, mobile QuickPass registration and card binding, unified users, and pan-user data, can be deduplicated and integrated into a cardholder relationship table. The fields include: bank card number, mobile phone number, ID card, device number, and Cloud QuickPass user number.

[0069] The preset time period may be set based on experience, for example, one week, one month, one quarter, etc.

[0070] Step 202: Analyze the transaction data of all cardholders within the preset time period to determine a plurality of cardholder relationship network communities.

[0071] Specifically, when obtaining multiple cardholder relationship network communities, the bank card number, mobile phone number, ID number, device number and application user number in the transaction data can be used as vertices, and the mobile phone number, ID number, device number and application user number in the same transaction data can be connected to the bank card number to obtain a cardholder relationship network diagram. Then, the cardholder relationship network diagram is divided into communities to obtain multiple cardholder relationship network communities. When dividing the communities, it is necessary to traverse each vertex according to the cardholder relationship network diagram, determine multiple cardholder relationship network diagrams with connected edges and set the community initial attribute value of each vertex, and then iteratively update the community initial attribute value of each vertex in each cardholder relationship network diagram with connected edges, until the community attribute values ​​of all vertices in each cardholder relationship network diagram with connected edges are changed to the minimum community attribute value in each cardholder relationship network diagram with connected edges, and the minimum community attribute value is determined as the community ID of each cardholder relationship network diagram with connected edges. Finally, the community ID is associated with the transaction data of each vertex in the cardholder relationship network graph with a connecting edge, and multiple cardholder relationship network communities are obtained.

[0072] For example, all fields in the cardholder relationship table are taken as vertices, and the fields in the same record are connected to the card number. Spark GraphX ​​technology is used to build a cardholder relationship network graph with hundreds of millions of vertices and hundreds of millions of edges (for example Figure 3 The full update can be done once a month.

[0073] Then based on Figure 3 The cardholder relationship network diagram shown can be divided into communities using the connected component algorithm and the Louvain algorithm. The specific steps are as follows:

[0074] Use the spark graphx technology to calculate the connected components of the whole graph. Initially, each vertex has a unique integer id value as the initial community attribute value. Each vertex only considers its own value and the values of all its neighboring vertices with connected edges. In each iteration, update its own value to the minimum of its own id value and the id values of all its first-order neighbors. The calculation process can be as Figure 4 shown. After the iteration converges, the community attribute values of all vertices in the same community will become the minimum id value of the vertices in the community, and this is used as the community ID of the cardholder relationship network community.

[0075] Among them, due to the large amount of data, memory overflow problems may occur during the calculation process. To solve this technical problem, when iteratively updating the initial community attribute values, specifically, first count the number of first-order neighbor vertices of each vertex. Then judge whether the number of first-order neighbor vertices is less than the preset data processing threshold. If so, record the first-order neighbor vertices of each vertex in the first set for iterative update of the initial community attribute values; otherwise, record the first-order neighbor vertices of each vertex in the second set and initialize them to the driver side; and broadcast the vertices in the second set to other thread pools (executors) so that other executors can obtain vertices from the second set for iterative update of the initial community attribute values. The preset data processing threshold can be determined according to the memory of the thread pool.

[0076] For example, as Figure 5 shown, the following steps can be carried out when performing iterative update:

[0077] 1. Count the number of first-order neighbor vertices cnt[D] of each vertex D (Data).

[0078] 2. Estimate the data processing threshold N according to the memory size of the executor (thread pool).

[0079] 3. If cnt[D] < N, record the vertex in set A; if cnt[D] >= N, record it in set B.

[0080] 4. The vertices in set A update their values through edge transmission; the vertices and values in set B are initialized to the driver side through sc.broadcast (broadcast variable) and broadcast to each executor, so that each executor only needs to obtain the values of the vertices in B from memory for update.

[0081] 5. Record the number of updated vertices as check;

[0082] 6. If check > 0, repeat steps 1 to 5; otherwise, the iteration ends and the result is output.

[0083] If Spark encounters memory resource problems during the calculation process, it will clear older non-persistent memory data to ensure the normal calculation process. In the process of large-scale data calculation, due to memory resource limitations, some useful temporary data is mistakenly cleared and needs to be recalculated. Therefore, a checkpoint is performed on the entire graph data in each iteration to ensure data quality and speed up the calculation.

[0084] Through the above strategies, the time of the connected component algorithm under large data volumes is reduced by 50% to 70% compared with the built-in algorithm in graphX, and the network transmission and memory consumption are also reduced to a certain extent.

[0085] Import the community ID obtained in the above steps into the cardholder relationship network library to generate a cardholder community table (i.e., the cardholder relationship network community), which contains three fields: attribute value, attribute type, and community ID. It supports community ID queries for any bank card, mobile phone number, ID card, device, or UnionPay APP user number. Among them, the attribute value is the value corresponding to the attribute type. For example, if the attribute type is bank card, the attribute value is the bank card number; if the attribute type is ID card, the attribute value is the ID card number.

[0086] The above community division method can divide the relationship network of 2.5 billion vertices and 6 billion edges into 390 million communities within 3 hours. All elements in the same community are regarded as a community of interests, including information of individuals, relatives and friends.

[0087] Step 203, for each cardholder relationship network community, determine the credit card consumption transaction data and debit card agency payment transaction data in the cardholder relationship network community; and determine abnormal data in the cardholder relationship network community based on the credit card consumption transaction data and the debit card agency payment transaction data.

[0088] The abnormal data is the data generated by abnormal use of credit cards.

[0089] For each cardholder relationship network community, one-to-one transaction matching and many-to-one transaction matching can be performed, as follows:

[0090] One-to-one transaction matching:

[0091] The credit card consumption transaction data and debit card payment transaction data in the cardholder relationship network community whose percentage of amount difference is less than a first threshold and whose absolute value of transaction time difference is less than a second threshold are determined, and the credit card consumption transaction data and debit card payment transaction data whose percentage of amount difference is less than the first threshold and whose absolute value of transaction time difference is less than the second threshold are determined as abnormal data in the cardholder relationship network community.

[0092] Many-to-one transaction matching:

[0093] After removing the credit card consumption transaction data and debit card payment transaction data whose percentage of amount difference is less than a first threshold and whose absolute value of transaction time difference is less than a second threshold from the credit card consumption transaction data and debit card payment transaction data in the cardholder relationship network community, traverse each payment transaction data to obtain its corresponding community ID, payment amount and payment time.

[0094] The consumption transaction data of each credit card under the community ID is traversed to obtain a transaction set; a transaction set subset with the smallest absolute value of the difference between the sum of the transaction amounts and the amount paid on behalf of others is determined from the transaction set; if the percentage of the absolute value of the difference between the sum of the transaction amounts and the amount paid on behalf of others in the transaction set subset is less than a first threshold and the difference between the transaction time and the time paid on behalf of others in the transaction set subset is less than a second threshold, the consumption transaction data of the credit card in the transaction set subset and the traversed payment on behalf of others transaction data are determined as abnormal data in the cardholder relationship network community.

[0095] The first threshold and the second threshold can be set based on experience.

[0096] Specifically, based on the business feature of "different cards in the same community make transactions of consumption first and payment later at different merchants within a short period of time, and the transaction amounts are similar", information processing can be performed, specifically:

[0097] Credit card consumption transaction data and debit card payment transaction data are determined from the cardholder relationship network community.

[0098] One-to-one matching: traverse the credit card consumption transactions and debit card payment transactions in the same community. If the percentage difference between the amounts of the two transactions is less than the threshold r (the first threshold), and the absolute value of the transaction time difference is less than the threshold T (the second threshold), the two transactions are marked as one-to-one abnormal usage matching transactions.

[0099] Many-to-one matching:

[0100] The consumption transactions and payment transactions detected in the one-to-one matching are removed from the cardholder relationship network community, and then each payment transaction trans1 is traversed to obtain its community ID and payment amount trans_at1, and payment time time1; each credit card under the community ID is traversed, and the set of all consumption transactions of the credit card is recorded as trans_set2. The backtracking + limit method is used to filter out a subset trans_set2_sub in trans_set2 so that the absolute value of the difference between the sum of the transaction amounts in trans_set2_sub and trans_at1 is minimized, and trans_at_set2_sub is recorded as the set of all transaction amounts in trans_set2_sub.

[0101] If the following two conditions are met:

[0102] ①、|sum(trans_at_set2_sub)-trans_at1| / sum(trans_at_set2_sub) <r;

[0103] ②, the maximum value of the time difference between two transactions in trans_at_set2_sub and trans1 <T;

[0104] Then match = (trans_at_set2_sub, trans1) is a many-to-one abnormal use pair transaction.

[0105] Since there is no polynomial time algorithm, the embodiment of the present invention controls the number of recursions to 1 billion times by adopting the backtracking + bounding method. The entire algorithm is implemented using MR, which can complete the abnormal data detection of about 3 billion transaction data in a week within 2 hours. Taking June XX to June XX, 2019 as an example, the detection results are as follows: 376 acquiring banks, 4959103 merchants, and 7028480 communities are suspected of abnormal use; 23685132 consumer transactions with suspected abnormal data, with an amount of 100 billion yuan; 25736050 payment transactions with suspected abnormal use, with an amount of 110 billion yuan. The model results have been recognized by the business operation center.

[0106] The above embodiment shows that the transaction data of all cardholders within a preset period of time is obtained, the transaction data of all cardholders within the preset period of time is analyzed, and multiple cardholder relationship network communities are determined. For each cardholder relationship network community, the credit card consumption transaction data and debit card payment transaction data in the cardholder relationship network community are determined, and the abnormal data in the cardholder relationship network community is determined based on the credit card consumption transaction data and the debit card payment transaction data. By obtaining the transaction data of all cardholders of each commercial bank within a preset period of time, and dividing each transaction data into communities based on graph calculation, the transaction data belonging to the same community is obtained, and then the transaction data of the same community is detected for abnormal data, the accuracy and coverage of abnormal data detection can be improved, and the efficiency of abnormal data detection of the system can be improved, and the power consumption of the system can be reduced.

[0107] Based on the same technical concept, Figure 6 The structure of an information processing device provided by an embodiment of the present invention is exemplarily shown. The device can execute the information processing process. Figure 1 The server 100 shown in may also be the server 100.

[0108] like Figure 6 As shown, the device specifically includes:

[0109] The acquisition unit 601 is used to acquire the transaction data of all cardholders within a preset period of time;

[0110] The processing unit 602 is used to analyze the transaction data of all cardholders within the preset time period to determine multiple cardholder relationship network communities; for each cardholder relationship network community, determine the credit card consumption transaction data and debit card payment transaction data in the cardholder relationship network community; and determine abnormal data in the cardholder relationship network community based on the credit card consumption transaction data and the debit card payment transaction data.

[0111] Optionally, the processing unit 602 is specifically configured to:

[0112] Obtaining transaction data of all cardholders within the preset time period; the transaction data includes bank card number, mobile phone number, ID card number, device number and application user number;

[0113] The bank card number, mobile phone number, ID number, device number and application user number in the transaction data are used as vertices, and the mobile phone number, ID number, device number and application user number in the same transaction data are connected to the bank card number to obtain a cardholder relationship network diagram;

[0114] The cardholder relationship network diagram is divided into communities to obtain a plurality of cardholder relationship network communities.

[0115] Optionally, the processing unit 602 is specifically configured to:

[0116] According to the cardholder relationship network diagram, traverse each vertex, determine multiple cardholder relationship network diagrams with connecting edges and set the community initial attribute value of each vertex;

[0117] Iteratively updating the community initial attribute value of each vertex in each cardholder relationship network graph with connected edges until the community attribute values ​​of all vertices in each cardholder relationship network graph with connected edges are changed to the minimum community attribute value in each cardholder relationship network graph with connected edges, and determining the minimum community attribute value as the community ID of each cardholder relationship network graph with connected edges; wherein each iterative update is to update the community attribute value of the vertex to the minimum value of the community attribute values ​​of all first-order neighbor vertices and its own community attribute value;

[0118] The community ID is associated with the transaction data of each vertex in the cardholder relationship network graph having a connecting edge, so as to obtain the plurality of cardholder relationship network communities.

[0119] Optionally, the processing unit 602 is specifically configured to:

[0120] Count the number of first-order neighbor vertices of each vertex;

[0121] If the number of the first-order neighbor vertices is less than a preset data processing threshold, the first-order neighbor vertices of each vertex are recorded in the first set for iterative update of the initial attribute value of the community; otherwise, the first-order neighbor vertices of each vertex are recorded in the second set and initialized to the driver; and the vertices in the second set are broadcast to other thread pool executors, so that the other executors obtain vertices from the second set for iterative update of the initial attribute value of the community.

[0122] Optionally, the processing unit 602 is specifically configured to:

[0123] Determining, from the credit card consumption transaction data and debit card agency payment transaction data in the cardholder relationship network community, the credit card consumption transaction data and debit card agency payment transaction data whose percentage of amount difference is less than a first threshold and whose absolute value of transaction time difference is less than a second threshold, and determining the credit card consumption transaction data and debit card agency payment transaction data whose percentage of amount difference is less than the first threshold and whose absolute value of transaction time difference is less than the second threshold as abnormal data in the cardholder relationship network community;

[0124] After removing the credit card consumption transaction data and debit card agency payment transaction data whose percentage of the amount difference is less than the first threshold and whose absolute value of the transaction time difference is less than the second threshold from the credit card consumption transaction data and debit card agency payment transaction data in the cardholder relationship network community, traverse each agency payment transaction data to obtain its corresponding community ID, agency payment amount and agency payment time;

[0125] Traverse the consumption transaction data of each credit card under the community ID to obtain a transaction set;

[0126] Determine from the transaction set a transaction set subset having the smallest absolute value of the difference between the sum of the transaction amounts and the payment amount;

[0127] If the percentage of the absolute value of the difference between the sum of the transaction amounts of the transaction set subset and the payment amount is less than the first threshold and the difference between the transaction time and the payment time in the transaction set subset is less than the second threshold, the consumption transaction data of the credit card in the transaction set subset and the traversed payment transaction data are determined as abnormal data in the cardholder relationship network community.

[0128] Based on the same technical concept, an embodiment of the present invention further provides a computing device, including:

[0129] A memory for storing program instructions;

[0130] The processor is used to call the program instructions stored in the memory and execute the above information processing method according to the obtained program.

[0131] Based on the same technical concept, an embodiment of the present invention further provides a computer-readable non-volatile storage medium, including computer-readable instructions. When a computer reads and executes the computer-readable instructions, the computer executes the above-mentioned information processing method.

[0132] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0133] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture including an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0134] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process in the computer or other programmable device. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0135] Although the preferred embodiments of the present invention have been described, those skilled in the art may make other changes and modifications to these embodiments once they have learned the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications that fall within the scope of the present invention.

[0136] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include these modifications and variations.

Claims

1. A method of information processing, characterized in that: include: Obtain transaction data of all cardholders within a preset period; Analyzing the transaction data of all cardholders within the preset time period to determine a plurality of cardholder relationship network communities; For each cardholder relationship network community, determining the credit card consumption transaction data and debit card payment transaction data in the cardholder relationship network community; Determining abnormal data in the cardholder relationship network community based on the credit card consumption transaction data and the debit card agency payment transaction data; The analysis of the transaction data of all cardholders within the preset period of time to determine a plurality of cardholder relationship network communities includes: Obtaining transaction data of all cardholders within the preset time period; the transaction data includes bank card number, mobile phone number, ID card number, device number and application user number; The bank card number, mobile phone number, ID number, device number and application user number in the transaction data are taken as vertices, and the mobile phone number, ID number, device number and application user number in the same transaction data are connected to the bank card number to obtain a cardholder relationship network diagram; Dividing the cardholder relationship network graph into communities to obtain a plurality of cardholder relationship network communities; Determining abnormal data in the cardholder relationship network community based on the credit card consumption transaction data and the debit card payment transaction data includes: Determining, from the credit card consumption transaction data and debit card agency payment transaction data in the cardholder relationship network community, the credit card consumption transaction data and debit card agency payment transaction data whose percentage of amount difference is less than a first threshold and whose absolute value of transaction time difference is less than a second threshold, and determining the credit card consumption transaction data and debit card agency payment transaction data whose percentage of amount difference is less than the first threshold and whose absolute value of transaction time difference is less than the second threshold as abnormal data in the cardholder relationship network community; After removing the credit card consumption transaction data and debit card agency payment transaction data whose percentage of the amount difference is less than the first threshold and whose absolute value of the transaction time difference is less than the second threshold from the credit card consumption transaction data and debit card agency payment transaction data in the cardholder relationship network community, traverse each agency payment transaction data to obtain its corresponding community ID, agency payment amount and agency payment time; Traverse the consumption transaction data of each credit card under the community ID to obtain a transaction set; By backtracking and limiting, a transaction set subset having the smallest absolute value of the difference between the sum of the transaction amounts and the payment amount is determined from the transaction set; If the percentage of the absolute value of the difference between the sum of the transaction amounts of the transaction set subset and the payment amount is less than the first threshold and the difference between the transaction time and the payment time in the transaction set subset is less than the second threshold, the consumption transaction data of the credit card and the traversed payment transaction data in the transaction set subset are determined as abnormal data in the cardholder relationship network community.

2. The method according to claim 1, characterized in that The cardholder relationship network diagram is divided into communities to obtain a plurality of cardholder relationship network communities, including: According to the cardholder relationship network diagram, traverse each vertex, determine multiple cardholder relationship network diagrams with connecting edges and set the community initial attribute value of each vertex; Iteratively updating the community initial attribute value of each vertex in each cardholder relationship network graph with connected edges until the community attribute values ​​of all vertices in each cardholder relationship network graph with connected edges are changed to the minimum community attribute value in each cardholder relationship network graph with connected edges, and determining the minimum community attribute value as the community ID of each cardholder relationship network graph with connected edges; wherein each iterative update is to update the community attribute value of the vertex to the minimum value of the community attribute values ​​of all first-order neighbor vertices and its own community attribute value; The community ID is associated with the transaction data of each vertex in the cardholder relationship network graph having a connecting edge, so as to obtain the plurality of cardholder relationship network communities.

3. The method according to claim 2, characterized in that The iterative updating of the community initial attribute value of each vertex in each cardholder relationship network graph having a connection edge includes: Count the number of first-order neighbor vertices of each vertex; If the number of the first-order neighbor vertices is less than a preset data processing threshold, the first-order neighbor vertices of each vertex are recorded in the first set for iterative update of the initial attribute value of the community; otherwise, the first-order neighbor vertices of each vertex are recorded in the second set and initialized to the driver; and the vertices in the second set are broadcast to other thread pool executors, so that the other thread pool executors obtain vertices from the second set for iterative update of the initial attribute value of the community.

4. An information processing device, characterized in that: include: An acquisition unit, used to acquire transaction data of all cardholders within a preset period of time; A processing unit, configured to analyze the transaction data of all cardholders within the preset period of time and determine a plurality of cardholder relationship network communities; For each cardholder relationship network community, determining the credit card consumption transaction data and the debit card agency payment transaction data in the cardholder relationship network community; determining the abnormal data in the cardholder relationship network community based on the credit card consumption transaction data and the debit card agency payment transaction data; Wherein, the processing unit is specifically used for: Obtaining transaction data of all cardholders within the preset time period; the transaction data includes bank card number, mobile phone number, ID card number, device number and application user number; The bank card number, mobile phone number, ID number, device number and application user number in the transaction data are taken as vertices, and the mobile phone number, ID number, device number and application user number in the same transaction data are connected to the bank card number to obtain a cardholder relationship network diagram; Dividing the cardholder relationship network graph into communities to obtain a plurality of cardholder relationship network communities; Determining, from the credit card consumption transaction data and debit card agency payment transaction data in the cardholder relationship network community, the credit card consumption transaction data and debit card agency payment transaction data whose percentage of amount difference is less than a first threshold and whose absolute value of transaction time difference is less than a second threshold, and determining the credit card consumption transaction data and debit card agency payment transaction data whose percentage of amount difference is less than the first threshold and whose absolute value of transaction time difference is less than the second threshold as abnormal data in the cardholder relationship network community; After removing the credit card consumption transaction data and debit card agency payment transaction data whose percentage of the amount difference is less than the first threshold and whose absolute value of the transaction time difference is less than the second threshold from the credit card consumption transaction data and debit card agency payment transaction data in the cardholder relationship network community, traverse each agency payment transaction data to obtain its corresponding community ID, agency payment amount and agency payment time; Traverse the consumption transaction data of each credit card under the community ID to obtain a transaction set; By backtracking and limiting, a transaction set subset having the smallest absolute value of the difference between the sum of the transaction amounts and the payment amount is determined from the transaction set; If the percentage of the absolute value of the difference between the sum of the transaction amounts of the transaction set subset and the payment amount is less than the first threshold and the difference between the transaction time and the payment time in the transaction set subset is less than the second threshold, the consumption transaction data of the credit card and the traversed payment transaction data in the transaction set subset are determined as abnormal data in the cardholder relationship network community.

5. The device according to claim 4, characterized in that The processing unit is specifically used for: According to the cardholder relationship network diagram, traverse each vertex, determine multiple cardholder relationship network diagrams with connecting edges and set the community initial attribute value of each vertex; Iteratively updating the community initial attribute value of each vertex in each cardholder relationship network graph with connected edges until the community attribute values ​​of all vertices in each cardholder relationship network graph with connected edges are changed to the minimum community attribute value in each cardholder relationship network graph with connected edges, and determining the minimum community attribute value as the community ID of each cardholder relationship network graph with connected edges; wherein each iterative update is to update the community attribute value of the vertex to the minimum value of the community attribute values ​​of all first-order neighbor vertices and its own community attribute value; The community ID is associated with the transaction data of each vertex in the cardholder relationship network graph having a connecting edge, so as to obtain the plurality of cardholder relationship network communities.

6. The device according to claim 5, characterized in that The processing unit is specifically used for: Count the number of first-order neighbor vertices of each vertex; If the number of the first-order neighbor vertices is less than a preset data processing threshold, the first-order neighbor vertices of each vertex are recorded in the first set for iterative update of the initial attribute value of the community; otherwise, the first-order neighbor vertices of each vertex are recorded in the second set and initialized to the driver; and the vertices in the second set are broadcast to other thread pool executors, so that the other thread pool executors obtain vertices from the second set for iterative update of the initial attribute value of the community.

7. A computing device, characterized in that include: A memory for storing program instructions; A processor, configured to call the program instructions stored in the memory, and execute the method according to any one of claims 1 to 3 according to the obtained program.

8. A computer-readable non-volatile storage medium, characterized in that: The method comprises computer-readable instructions, and when a computer reads and executes the computer-readable instructions, the computer is caused to execute the method according to any one of claims 1 to 3.

Images