Risk identification method and device based on alliance chain

Abstract

Embodiments of the present invention disclose a consortium chain-based risk identification method and device, relating to the field of blockchain technology. The method comprises: obtaining all bank transfer records of a target customer from a consortium chain system, wherein the consortium chain system includes a bank node corresponding to each bank, and each bank uploads the bank transfer records of its own customers to the consortium chain system through the corresponding bank node; obtaining a first risk identification result obtained by performing risk identification on all bank transfer records of the target customer using each bank's respective first risk identification model; combining all bank transfer records of the target customer with the first risk identification result, and then inputting the result into a preset second risk identification model to obtain a second risk identification result for the target customer. The present invention helps improve the accuracy of risk identification.

Description

Technical Field

[0001] The present invention relates to the field of blockchain technology, and in particular to a risk identification method and device based on an alliance chain. Background Art

[0002] In current production applications, bank transfer records and account information are stored on individual bank data center hosts or data server containers, making cross-bank fund tracing difficult. Currently, user account information and transfer information are stored on individual bank data center hosts or data servers. This data storage structure presents two issues: 1. Internal records within each bank are difficult to share across banks, making zero-trust data sharing difficult to achieve. 2. Interbank transfer records are subject to tampering when retrieved.

[0003] Currently, banks rely solely on internal data and proprietary algorithms for risk identification, which inevitably leads to inaccuracies. Improving the accuracy of bank risk identification is a pressing technical challenge. Summary of the Invention

[0004] In order to solve at least one technical problem in the above-mentioned background technology, the present invention proposes a risk identification method and device based on alliance chain.

[0005] To achieve the above objectives, according to one aspect of the present invention, a risk identification method based on a consortium chain is provided, the method comprising:

[0006] Obtain all bank transfer records of the target customer from the consortium chain system, wherein the consortium chain system includes a bank node corresponding to each bank, and each bank uploads the bank transfer records of its own customers to the consortium chain system through the corresponding bank node;

[0007] Obtaining a first risk identification result obtained by performing risk identification on all bank transfer records of the target customer using a first risk identification model of each bank, wherein the first risk identification model of each bank is trained based on the bank transfer records of customers within the bank;

[0008] All bank transfer records of the target customer are combined with the first risk identification result, and then input into a preset second risk identification model to obtain a second risk identification result of the target customer, wherein the second risk identification model is obtained by training a preset machine learning model based on training samples, and the training samples include bank transfer records used for model training and the first risk identification results corresponding to the bank transfer records used for model training, and the training samples are marked with risk identification results.

[0009] Optionally, each bank also uploads the bank account information of its own customers to the alliance chain system through the corresponding bank node;

[0010] The acquisition of all bank transfer records of target customers from the alliance chain system specifically includes:

[0011] Match the target customer's customer information with all bank account information in the alliance chain system to determine all bank accounts corresponding to the target customer;

[0012] All bank transfer records corresponding to all bank accounts are obtained from the alliance chain system.

[0013] Optionally, obtaining a first risk identification result obtained by performing risk identification on all bank transfer records of the target customer using each bank's respective first risk identification model specifically includes:

[0014] Obtain each bank's respective first risk identification model;

[0015] All bank transfer records of the target customer are input into the first risk identification model of each bank respectively, and the first risk identification results output by each first risk identification model are obtained.

[0016] Optionally, the risk identification method based on the alliance chain further includes:

[0017] The second risk identification result corresponding to the target customer is uploaded to the alliance chain system.

[0018] Optionally, the risk identification method based on the alliance chain further includes:

[0019] Obtaining the training sample;

[0020] The preset machine learning model is trained according to the training samples to obtain the second risk identification model.

[0021] Optionally, obtaining the training sample specifically includes:

[0022] Obtain bank transfer records for model training;

[0023] Inputting the bank transfer record used for model training into each of the first risk identification models to obtain a first risk identification result corresponding to the bank transfer record used for model training;

[0024] The bank transfer record used for model training is combined with the corresponding first risk identification result, and then labeled with the risk identification result to obtain the training sample.

[0025] Optionally, obtaining a first risk identification result obtained by performing risk identification on all bank transfer records of the target customer using each bank's respective first risk identification model specifically includes:

[0026] Sending a risk identification request to the bank node of each bank, so that each bank first obtains all bank transfer records of the target customer from the alliance chain system according to the risk identification request, then uses its own first risk identification model to perform risk identification on the obtained bank transfer records to obtain a first risk identification result, and finally uploads the obtained first risk identification result to the alliance chain system through the corresponding bank node;

[0027] The first risk identification results output by each first risk identification model are obtained from the alliance chain system.

[0028] To achieve the above objectives, according to another aspect of the present invention, a risk identification device based on a consortium chain is provided, the device comprising:

[0029] A bank transfer record acquisition unit, configured to acquire all bank transfer records of a target customer from a consortium chain system, wherein the consortium chain system includes a bank node corresponding to each bank, and each bank uploads the bank transfer records of its own customers to the consortium chain system through the corresponding bank node;

[0030] a first risk identification result acquisition unit, configured to acquire a first risk identification result obtained by performing risk identification on all bank transfer records of the target customer using a first risk identification model of each bank, wherein the first risk identification model of each bank is trained based on the bank transfer records of customers within the bank;

[0031] A risk identification unit is used to combine all bank transfer records of the target customer with the first risk identification result, and then input them into a preset second risk identification model to obtain a second risk identification result for the target customer, wherein the second risk identification model is obtained by training a preset machine learning model based on training samples, and the training samples include bank transfer records used for model training and the first risk identification result corresponding to the bank transfer records used for model training, and the training samples are marked with risk identification results.

[0032] In order to achieve the above-mentioned purpose, according to another aspect of the present invention, a computer device is further provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps of the above-mentioned risk identification method based on the alliance chain when executing the computer program.

[0033] In order to achieve the above-mentioned purpose, according to another aspect of the present invention, a computer-readable storage medium is further provided, on which a computer program / instruction is stored. When the computer program / instruction is executed by a processor, the steps of the above-mentioned risk identification method based on the alliance chain are implemented.

[0034] In order to achieve the above-mentioned purpose, according to another aspect of the present invention, a computer program product is further provided, including a computer program / instruction, which implements the steps of the above-mentioned risk identification method based on the alliance chain when executed by a processor.

[0035] The beneficial effects of the present invention are:

[0036] The present invention enables banks to form an alliance chain system, and each bank uploads the bank transfer records of its own customers to the alliance chain system through the corresponding bank node, thereby solving the problem of insufficient recognition accuracy due to the small amount of data of a single bank and the problem of tampering that may be faced by intra-bank data sharing. In addition, when performing risk identification, the present invention uses the first risk identification result obtained by risk identification using each bank's respective first risk identification model as a parameter for further risk identification, further improving the accuracy of risk identification and solving the problem of insufficient model identification accuracy due to the small amount of data of a single bank. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative work. In the drawings:

[0038] Figure 1 This is the first flow chart of the risk identification method based on the alliance chain according to an embodiment of the present invention;

[0039] Figure 2 This is a second flow chart of the risk identification method based on the alliance chain according to an embodiment of the present invention;

[0040] Figure 3 This is the third flow chart of the risk identification method based on the alliance chain according to an embodiment of the present invention;

[0041] Figure 4 This is a fourth flow chart of the risk identification method based on the alliance chain according to an embodiment of the present invention;

[0042] Figure 5 This is a fifth flow chart of the risk identification method based on the alliance chain according to an embodiment of the present invention;

[0043] Figure 6This is the sixth flow chart of the risk identification method based on the alliance chain according to an embodiment of the present invention;

[0044] Figure 7 This is a schematic diagram of the alliance chain system of the present invention;

[0045] Figure 8 It is a schematic diagram of the block structure of the present invention;

[0046] Figure 9 It is the overall flow chart of identification of the present invention;

[0047] Figure 10 This is a structural block diagram of a risk identification device based on a consortium chain according to an embodiment of the present invention;

[0048] Figure 11 Schematic diagram of a computer device according to an embodiment of the present invention. DETAILED DESCRIPTION

[0049] In order to enable those skilled in the art to better understand the solutions of the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the embodiments described are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts should fall within the scope of protection of the present invention.

[0050] It will be understood by those skilled in the art that embodiments of the present invention may be provided as methods, systems, or computer program products. Thus, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware. Furthermore, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to magnetic disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0051] It should be noted that the terms "including" and "having" and any variations thereof in the specification and claims of the present invention and the above-mentioned drawings are intended to cover non-exclusive inclusions. For example, a process, method, system, product or apparatus comprising a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or are inherent to these processes, methods, products or apparatuses.

[0052] It should be noted that, in the absence of conflict, the embodiments and features of the embodiments of the present invention can be combined with each other. The present invention will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.

[0053] It should be noted that the acquisition, storage, use, and processing of data in the technical solution of this application comply with the relevant provisions of national laws and regulations.

[0054] It should be noted that the risk identification method and device based on the alliance chain of the present invention can be used in the financial field, and can also be used in any field other than the financial field. The application field of the risk identification method and device based on the alliance chain of the present invention is not limited.

[0055] The present invention addresses the current information contradiction between transfer information sharing and security, as well as the problem of insufficient bank risk identification efficiency, by combining blockchain consortium chain technology with deep learning technology to provide a traceable transfer record and shared risk identification method based on a consortium chain. The present invention modifies blockchain block nodes, customizes block classification, and constructs a transfer information sharing consortium chain, making it more suitable for storing bank transfer information and account information, and more convenient for data application. The present invention overcomes the contradiction between cross-bank sharing and security of transfer information caused by the current storage method of bank transfer records. Based on the data integrity and data tamper-proofing of the consortium chain, the ability to register access records ensures that transfer information is authentic and not tampered with, while enabling convenient traceability of cross-bank transfers and facilitating the sharing of transfer information and user information. The present invention overcomes the difficulty of current bank risk identification methods that can only use the bank's own data and risk identification algorithms. Based on blockchain technology, the present invention obtains data from all participating banks, performs identification processing on a user-by-user basis, and simultaneously shares risk identification algorithms. The most accurate identification results are obtained based on neural networks, making risk identification more precise.

[0056] The present invention establishes a consortium chain system, the specific structure can be as follows Figure 7 As shown, the alliance chain system includes the bank nodes corresponding to each bank ( Figure 7 The traceable bank transaction alliance chain (the bank-side chain) and the risk calculation and supervision node are connected. Each bank node uploads the bank account information and bank transfer records of its own customers to the alliance chain. The risk calculation and supervision node performs risk identification based on the data on the alliance chain, obtains bank account risk information (i.e., the second risk identification result of the present invention), and uploads the obtained bank account risk information to the alliance chain. As a result, each bank can obtain more accurate risk identification results from the alliance chain.

[0057] In one embodiment of the present invention, Docker containers can be used as consortium chain nodes using Docker technology to facilitate capacity expansion. Each participating bank in the consortium chain maintains several blockchain nodes, overseeing each other. Meanwhile, a third-party regulatory body maintains several risk calculation and regulatory nodes.

[0058] Figure 8 This is a block structure diagram of the present invention, such as Figure 8As shown, the data blocks on the alliance chain of the present invention specifically include: a core block of bank transfer records, a bank account information block, an access record storage block, and a bank account risk information block. The block of the present invention includes three areas: a block header, a hash storage area, and a data storage area. The block header stores the timestamp, the generated Merkle tree root, the block hash calculated by the current Merkle tree root and the hash value of the previous block, the UUID of the node, and the node category. The hash storage area stores the Merkle tree structure composed of leaf nodes with the hash value of the information stored in the current block. The data storage area stores the data information stored in the block in the form of KV, such as transfer records, account information, etc.

[0059] The core block of bank transfer records stores bank transfer record information in the data storage area. The block Merkle tree is constructed based on the hash value of the bank transfer record information, and then the session core block is constructed.

[0060] The bank account information block, the data storage area stores the user's bank account information. Based on the bank account information hash value, a Merkle tree is constructed to build the access record storage block.

[0061] Bank account risk information block, the risk calculation and supervision node calculates the user account risk information, stores the user account risk information in the data storage area, builds a Merkle tree based on the hash value of the user account risk information, and then builds the bank account risk information block.

[0062] like Figure 8 As shown, the present invention constructs a consortium chain by following the above blocks in the order they are constructed, and creates a block index based on the block type. The block index indicates the location of data in the consortium chain and is used to quickly find data on the consortium chain. In one embodiment of the present invention, the block index can be stored on the consortium chain or in other data storage spaces. The block index specifically includes: a core block index for bank transfer records, a block index for bank account information, a block index for access record storage, and a block index for bank account risk information.

[0063] It should be noted that the risk identification method based on the alliance chain in the following embodiments of the present invention is applied to the risk calculation and supervision node, that is, the executor of the risk identification method based on the alliance chain in the following embodiments of the present invention is the risk calculation and supervision node.

[0064] Figure 1 This is the first flow chart of the risk identification method based on the alliance chain according to an embodiment of the present invention. Figure 1 As shown, in one embodiment of the present invention, the risk identification method based on the alliance chain of the present invention includes steps S101 to S103.

[0065] Step S101: Obtain all bank transfer records of the target customer from the alliance chain system, wherein the alliance chain system includes a bank node corresponding to each bank, and each bank uploads the bank transfer records of its own customers to the alliance chain system through the corresponding bank node.

[0066] In the present invention, each bank uploads the bank transfer records of its own customers to the alliance chain through the corresponding bank node, thereby realizing the sharing of bank transfer records of customers within each bank, and the use of blockchain technology also prevents information tampering.

[0067] Step S102, obtaining a first risk identification result obtained by performing risk identification on all bank transfer records of the target customer using each bank's respective first risk identification model, wherein the first risk identification model of each bank is trained based on the bank transfer records of customers within the bank.

[0068] In the present invention, each bank has its own risk identification model, namely, the first risk identification model. However, each bank's first risk identification model is trained based on in-bank data, which can lead to inaccuracy due to insufficient data. The present invention uses the identification results of each bank's first risk identification model as a parameter for further risk identification, helping to improve the accuracy of risk identification.

[0069] The number of first risk identification results obtained in this step is consistent with the number of participating banks in the alliance chain system.

[0070] Step S103: Combine all bank transfer records of the target customer with the first risk identification result, and then input them into a preset second risk identification model to obtain a second risk identification result of the target customer, wherein the second risk identification model is obtained by training a preset machine learning model based on training samples, and the training samples include bank transfer records used for model training and the first risk identification result corresponding to the bank transfer records used for model training, and the training samples are marked with risk identification results.

[0071] The present invention pre-trains a second risk identification model and sets the second risk identification model in a risk calculation and supervision node of the alliance chain system.

[0072] It can be seen that the present invention enables each bank to form an alliance chain system, and each bank uploads the bank transfer records of its own customers to the alliance chain system through the corresponding bank node, thereby solving the problem of insufficient recognition accuracy due to the small amount of data of a single bank and the problem of tampering that may be faced by intra-bank data sharing. In addition, when performing risk identification, the present invention uses the first risk identification result obtained by risk identification using each bank's respective first risk identification model as a parameter for further risk identification, further improving the accuracy of risk identification and solving the problem of insufficient model identification accuracy due to the small amount of data of a single bank.

[0073] In the present invention, each bank uploads the bank transfer records of its own customers to the alliance chain system through the corresponding bank node, thereby improving the accuracy and completeness of user transfer record tracing. In the user transfer record tracing scenario, when one of the participating banks initiates transfer tracing, the system accesses the alliance chain to obtain the two parties to the transaction. If it is found that the funds in the fund receiving account are approximately less than the transfer amount, it will continue to search for transaction records after the transaction time, with the fund receiving account as the fund transfer account. Ensure that the sum of the account balance and the found transfer amount is greater than the previous transfer amount. By filtering in this way until the tracing is initiated, a complete fund transaction chain can be quickly obtained, facilitating fund tracing. At the same time, based on the ID number of the account user, the user's deposit status in all banks can be obtained, facilitating the cross-bank deduction and freezing of funds.

[0074] In one embodiment of the present invention, each bank also uploads the bank account information of its own customers to the alliance chain system through the corresponding bank node.

[0075] like Figure 2 As shown, the above step S101 of obtaining all bank transfer records of the target customer from the alliance chain system specifically includes steps S201 and S202.

[0076] Step S201: Match the customer information of the target customer with all bank account information in the alliance chain system to determine all bank accounts corresponding to the target customer.

[0077] Step S202: Obtain all bank transfer records corresponding to all bank accounts from the alliance chain system.

[0078] The present invention pre-establishes a block index that indicates the location of data in the consortium chain, allowing for rapid data retrieval from the consortium chain. The present invention can quickly retrieve bank account information from the blockchain using the bank account information block index, and can also quickly retrieve bank transfer records from the blockchain using the bank transfer record core block index.

[0079] like Figure 3 As shown, in one embodiment of the present invention, the acquisition of the above-mentioned step S102 is to obtain the first risk identification result obtained by using each bank's respective first risk identification model to perform risk identification on all bank transfer records of the target customer, specifically including step S301 and step S302.

[0080] Step S301: Obtain the first risk identification model of each bank.

[0081] Step S302: input all bank transfer records of the target customer into the first risk identification model of each bank respectively, and obtain the first risk identification results output by each first risk identification model.

[0082] In one embodiment of the present invention, the present invention pre-sets the first risk identification model of each bank in the risk calculation and supervision node of the alliance chain system, and the first risk identification result is specifically obtained by the risk calculation and supervision node based on each first risk identification model.

[0083] In one embodiment of the present invention, after the above step S103, the risk identification method based on the alliance chain of the present invention further includes:

[0084] The second risk identification result corresponding to the target customer is uploaded to the alliance chain system.

[0085] After obtaining the final risk identification result (i.e., the second risk identification result), the present invention broadcasts the information of the two parties to the transaction and the risk identification result within the alliance chain, and stores them in the bank account risk block in the alliance chain as the basis for calculating the first-level risk identification model.

[0086] like Figure 4 As shown, in one embodiment of the present invention, the risk identification method based on the alliance chain of the present invention includes step S401 and step S402.

[0087] Step S401: Obtain the training sample.

[0088] Step S402: Train the preset machine learning model according to the training samples to obtain the second risk identification model.

[0089] In one embodiment of the present invention, the second risk identification model is specifically obtained by training a residual neural network.

[0090] like Figure 5 As shown, in one embodiment of the present invention, the step S401 of obtaining the training sample specifically includes steps S501 to S503.

[0091] Step S501: Obtain bank transfer records for model training.

[0092] Step S502: input the bank transfer record used for model training into each of the first risk identification models to obtain a first risk identification result corresponding to the bank transfer record used for model training.

[0093] Step S503: The bank transfer record used for model training is combined with the corresponding first risk identification result, and then labeled with the risk identification result to obtain the training sample.

[0094] like Figure 6 As shown, in one embodiment of the present invention, the acquisition in the above step S102 uses each bank's respective first risk identification model to perform risk identification on all bank transfer records of the target customer, and the obtained first risk identification result specifically includes step S601 and step S602.

[0095] Step S601: Send a risk identification request to the bank node of each bank, so that each bank first obtains all bank transfer records of the target customer from the alliance chain system according to the risk identification request, and then uses its own first risk identification model to perform risk identification on the obtained bank transfer records to obtain a first risk identification result, and finally uploads the obtained first risk identification result to the alliance chain system through the corresponding bank node.

[0096] Step S602: Obtain the first risk identification results output by each first risk identification model from the alliance chain system.

[0097] In one embodiment of the present invention, the first risk identification model is set in each bank node, the first risk identification result is obtained by the bank node and uploaded to the alliance chain, and the risk calculation and supervision node obtains each first risk identification result from the alliance chain.

[0098] In this invention, all participating banks in the consortium chain deploy their own risk identification models (i.e., the first risk identification model) in the risk calculation and supervision nodes, achieving zero-trust full-chain data acquisition and calculation. This allows for individual risk identification based on user ID numbers as the primary key and on a per-user basis.

[0099] This invention includes a two-level risk identification model. The first-level risk identification model is the existing risk identification model of each bank (i.e., the first risk identification model). The second-level model is the transfer risk result identification model based on the residual neural network constructed in this patent (i.e., the second risk identification model), which integrates and verifies the risk identification results of each bank, thereby improving identification accuracy and data utilization.

[0100] During model training, the present invention screens all bank transfer records of some normal users and some criminals, including those caught committing telecom fraud, and constructs a training dataset and a test dataset based on the identification results of the risk identification model for all the selected users (i.e., the first risk identification results). Then, based on deep learning technology, the training dataset is used to train a residual neural network to obtain a transfer risk result identification model. The trained transfer risk result identification model is tested using the test set, and the model is repeatedly trained with updated training and datasets to ultimately obtain a complete transfer risk result identification model.

[0101] Risk identification scenarios such as Figure 9 As shown, the target customer is the current transferee. The customer ID information of both accounts is used to retrieve all transfer records for the customer, with the risk identification records serving as input parameter 1. Input parameter 1 is passed to the risk identification model for all rows (i.e., the first risk identification model) to obtain the identification result set as input parameter 2. Input parameters 1 and 2 are combined to obtain input parameter 3, which is then passed to the transfer risk result identification model (i.e., the second risk identification model) to obtain the final risk identification result.

[0102] After obtaining the final risk identification results, the present invention broadcasts the transaction parties' information and the risk identification results within the consortium chain and stores them in the bank account risk block within the consortium chain as the basis for calculating the first-level risk identification model. Furthermore, if the risk identification results exceed the threshold, the transaction is blocked.

[0103] It can be seen from the above embodiments that the method of the present invention achieves at least the following beneficial effects:

[0104] 1. The present invention improves the blockchain construction structure to make it better adapted to the storage, acquisition and application of user transfer information and user account information.

[0105] 2. This blockchain-based method for storing bank transfer records for traceability effectively resolves the conflict between cross-bank sharing and security. Leveraging the alliance blockchain's data integrity, tamper-proofing, and access record registration, it ensures that transfer information is authentic and tamper-proof. It also facilitates cross-bank transfer traceability and facilitates the sharing of transfer and user information.

[0106] 3. This invention overcomes the difficulty that current bank risk identification methods can only use the bank's own data and risk identification algorithms. It obtains data from all participating banks based on blockchain technology, performs identification processing on a user-by-user basis, and shares risk identification algorithms at the same time. It obtains the most accurate identification results based on neural networks, making risk identification more precise.

[0107] It should be noted that the steps shown in the flowcharts of the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions, and that, although a logical order is shown in the flowcharts, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0108] Based on the same inventive concept, an embodiment of the present invention also provides a risk identification device based on a consortium chain, which can be used to implement the risk identification method based on a consortium chain described in the above embodiment, as described in the following embodiment. Since the principle of solving the problem by the risk identification device based on a consortium chain is similar to that of the risk identification method based on a consortium chain, the embodiment of the risk identification device based on a consortium chain can refer to the embodiment of the risk identification method based on a consortium chain, and the repeated parts will not be repeated. As used below, the term "unit" or "module" can be a combination of software and / or hardware that implements a predetermined function. Although the device described in the following embodiment is preferably implemented in software, the implementation of hardware, or a combination of software and hardware, is also possible and conceived.

[0109] Figure 10 This is a structural diagram of a risk identification device based on a consortium chain according to an embodiment of the present invention. Figure 10 As shown, in one embodiment of the present invention, the risk identification device based on the alliance chain of the present invention includes:

[0110] A bank transfer record acquisition unit 1 is configured to acquire all bank transfer records of a target customer from a consortium chain system, wherein the consortium chain system includes a bank node corresponding to each bank, and each bank uploads the bank transfer records of its own customers to the consortium chain system through the corresponding bank node;

[0111] a first risk identification result acquisition unit 2, configured to acquire a first risk identification result obtained by performing risk identification on all bank transfer records of the target customer using a first risk identification model of each bank, wherein the first risk identification model of each bank is trained based on the bank transfer records of customers within the bank;

[0112] The risk identification unit 3 is used to combine all the bank transfer records of the target customer with the first risk identification result, and then input them into a preset second risk identification model to obtain a second risk identification result of the target customer, wherein the second risk identification model is obtained by training a preset machine learning model based on a training sample, and the training sample includes the bank transfer records used for model training and the first risk identification result corresponding to the bank transfer records used for model training, and the training sample is marked with the risk identification result.

[0113] In one embodiment of the present invention, each bank also uploads the bank account information of its own customers to the alliance chain system through the corresponding bank node.

[0114] In one embodiment of the present invention, the bank transfer record obtaining unit 1 specifically includes:

[0115] A bank account determination module, configured to match the target customer's customer information with all bank account information in the alliance chain system to determine all bank accounts corresponding to the target customer;

[0116] The transfer record query module is used to obtain all bank transfer records corresponding to all bank accounts from the alliance chain system.

[0117] In one embodiment of the present invention, the first risk identification result obtaining unit 2 specifically includes:

[0118] A model acquisition module is used to obtain the first risk identification model of each bank;

[0119] The identification module is used to input all bank transfer records of the target customer into the first risk identification model of each bank respectively, and obtain the first risk identification results output by each first risk identification model.

[0120] In one embodiment of the present invention, the risk identification device based on the alliance chain of the present invention further includes:

[0121] The risk identification result uploading unit is used to upload the second risk identification result corresponding to the target customer to the alliance chain system.

[0122] In one embodiment of the present invention, the risk identification device based on the alliance chain of the present invention further includes:

[0123] A training sample acquisition unit, configured to acquire the training sample;

[0124] A model training unit is used to train a preset machine learning model according to the training samples to obtain the second risk identification model.

[0125] In one embodiment of the present invention, the training sample acquisition unit specifically includes:

[0126] The bank transfer record acquisition module is used to obtain bank transfer records for model training;

[0127] a first risk identification result determination module, configured to input the bank transfer record used for model training into each of the first risk identification models to obtain a first risk identification result corresponding to the bank transfer record used for model training;

[0128] The data processing module is used to combine the bank transfer record used for model training with the corresponding first risk identification result, and then label the risk identification result to obtain the training sample.

[0129] In one embodiment of the present invention, the first risk identification result obtaining unit 2 specifically includes:

[0130] a risk identification request sending module, configured to send a risk identification request to the bank node of each bank, so that each bank first obtains all bank transfer records of the target customer from the consortium chain system according to the risk identification request, then uses its own first risk identification model to perform risk identification on the obtained bank transfer records to obtain a first risk identification result, and finally uploads the obtained first risk identification result to the consortium chain system through the corresponding bank node;

[0131] The on-chain data acquisition module is used to obtain the first risk identification results output by each first risk identification model from the alliance chain system.

[0132] In order to achieve the above object, according to another aspect of the present application, a computer device is also provided. Figure 11 As shown, the computer device includes a memory, a processor, a communication interface and a communication bus. The memory stores a computer program that can be run on the processor. When the processor executes the computer program, the steps in the above embodiment method are implemented.

[0133] The processor may be a central processing unit (CPU). The processor may also be other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, or a combination of the above chips.

[0134] Memory, as a non-transitory computer-readable storage medium, can be used to store non-transitory software programs, non-transitory computer-executable programs, and units, such as the corresponding program units in the above-described method embodiments of the present invention. The processor executes the non-transitory software programs, instructions, and modules stored in memory to perform various processor functions and work data processing, thereby implementing the methods in the above-described method embodiments.

[0135] The memory may include a program storage area and a data storage area, wherein the program storage area may store an operating system, an application required for at least one function; the data storage area may store data created by the processor, etc. In addition, the memory may include a high-speed random access memory, and may also include a non-volatile memory, such as at least one disk storage device, a flash memory device, or other non-volatile solid-state storage device. In some embodiments, the memory may optionally include a memory remotely located relative to the processor, and these remote memories may be connected to the processor via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0136] The one or more units are stored in the memory, and when executed by the processor, perform the method in the above embodiment.

[0137] The specific details of the above-mentioned computer device can be understood by referring to the corresponding descriptions and effects in the above-mentioned embodiments, and will not be repeated here.

[0138] To achieve the above objectives, according to another aspect of the present application, a computer-readable storage medium is provided, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed in a computer processor, the steps in the above-mentioned risk identification method based on the alliance chain are implemented. Those skilled in the art will understand that all or part of the processes in the above-mentioned embodiment methods can be implemented by instructing the relevant hardware through a computer program. The program can be stored in a computer-readable storage medium, and when the program is executed, it can include the processes of the embodiments of the above-mentioned methods. The storage medium can be a magnetic disk, an optical disk, a read-only memory (ROM), a random access memory (RAM), a flash memory, a hard disk drive (HDD), or a solid-state drive (SSD); the storage medium can also include a combination of the above-mentioned types of memory.

[0139] In order to achieve the above-mentioned purpose, according to another aspect of the present application, a computer program product is also provided, including a computer program / instruction, which implements the steps of the above-mentioned risk identification method based on the alliance chain when executed by a processor.

[0140] Obviously, those skilled in the art will appreciate that the modules or steps of the present invention described above can be implemented using a general-purpose computing device. They can be centralized on a single computing device or distributed across a network of multiple computing devices. Alternatively, they can be implemented using program code executable by a computing device, which can then be stored in a storage device and executed by the computing device. Alternatively, they can be fabricated into separate integrated circuit modules, or multiple modules or steps can be fabricated into a single integrated circuit module for implementation. Thus, the present invention is not limited to any specific combination of hardware and software.

[0141] The foregoing description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. Those skilled in the art will readily appreciate that various modifications and variations of the present invention are possible. Any modifications, equivalent substitutions, or improvements made within the spirit and principles of the present invention are intended to be within the scope of protection of the present invention.

Claims

1. A risk identification method based on alliance chain, characterized in that: include: Obtain all bank transfer records of the target customer from the consortium chain system, wherein the consortium chain system includes a bank node corresponding to each bank, and each bank uploads the bank transfer records of its own customers to the consortium chain system through the corresponding bank node; Obtaining a first risk identification result obtained by performing risk identification on all bank transfer records of the target customer using a first risk identification model of each bank, wherein the first risk identification model of each bank is trained based on the bank transfer records of customers within the bank; All bank transfer records of the target customer are combined with the first risk identification result, and then input into a preset second risk identification model to obtain a second risk identification result of the target customer, wherein the second risk identification model is obtained by training a preset machine learning model based on training samples, and the training samples include bank transfer records used for model training and the first risk identification results corresponding to the bank transfer records used for model training, and the training samples are marked with the first risk identification results.

2. The risk identification method based on alliance chain according to claim 1 is characterized in that: Each bank also uploads the bank account information of its own customers to the alliance chain system through the corresponding bank node; The acquisition of all bank transfer records of target customers from the alliance chain system specifically includes: Match the target customer's customer information with all bank account information in the alliance chain system to determine all bank accounts corresponding to the target customer; All bank transfer records corresponding to all bank accounts are obtained from the alliance chain system.

3. The risk identification method based on alliance chain according to claim 1 is characterized in that: Obtaining a first risk identification result obtained by performing risk identification on all bank transfer records of the target customer using each bank's respective first risk identification model, specifically including: Obtain each bank's respective first risk identification model; All bank transfer records of the target customer are input into the first risk identification model of each bank respectively, and the first risk identification results output by each first risk identification model are obtained.

4. The risk identification method based on alliance chain according to claim 1 is characterized in that: Also includes: The second risk identification result corresponding to the target customer is uploaded to the alliance chain system.

5. The risk identification method based on alliance chain according to claim 1 is characterized in that: Also includes: Obtaining the training sample; The preset machine learning model is trained according to the training samples to obtain the second risk identification model.

6. The risk identification method based on alliance chain according to claim 5 is characterized in that: The obtaining of the training sample specifically includes: Obtain bank transfer records for model training; Inputting the bank transfer record used for model training into each of the first risk identification models to obtain a first risk identification result corresponding to the bank transfer record used for model training; The bank transfer record used for model training is combined with the corresponding first risk identification result, and then labeled with the risk identification result to obtain the training sample.

7. The risk identification method based on alliance chain according to claim 1 is characterized in that: Obtaining a first risk identification result obtained by performing risk identification on all bank transfer records of the target customer using each bank's respective first risk identification model, specifically including: Sending a risk identification request to the bank node of each bank, so that each bank first obtains all bank transfer records of the target customer from the alliance chain system according to the risk identification request, then uses its own first risk identification model to perform risk identification on the obtained bank transfer records to obtain a first risk identification result, and finally uploads the obtained first risk identification result to the alliance chain system through the corresponding bank node; The first risk identification results output by each first risk identification model are obtained from the alliance chain system.

8. A risk identification device based on alliance chain, characterized in that: include: A bank transfer record acquisition unit, configured to acquire all bank transfer records of a target customer from a consortium chain system, wherein the consortium chain system includes a bank node corresponding to each bank, and each bank uploads the bank transfer records of its own customers to the consortium chain system through the corresponding bank node; a first risk identification result acquisition unit, configured to acquire a first risk identification result obtained by performing risk identification on all bank transfer records of the target customer using a first risk identification model of each bank, wherein the first risk identification model of each bank is trained based on the bank transfer records of customers within the bank; A risk identification unit is used to combine all bank transfer records of the target customer with the first risk identification result, and then input them into a preset second risk identification model to obtain a second risk identification result for the target customer, wherein the second risk identification model is obtained by training a preset machine learning model based on training samples, and the training samples include bank transfer records used for model training and the first risk identification result corresponding to the bank transfer records used for model training, and the training samples are marked with the first risk identification result.

9. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program / instruction stored thereon, characterized in that: When the computer program / instructions are executed by a processor, the steps of the method according to any one of claims 1 to 7 are implemented.

11. A computer program product comprising a computer program / instructions, characterized in that When the computer program / instructions are executed by a processor, the steps of the method according to any one of claims 1 to 7 are implemented.

Images