Method, device, electronic equipment and computer program product for processing financial risk

Abstract

The application discloses a financial risk processing method and device, electronic equipment and computer program product. It is related to the field of financial technology. The method comprises: obtaining financial data of a target user in a preset time period under the authorization of the target user, wherein the financial data refers to business data associated with the target user when the target user handles financial business; extracting a risk identification model from a block chain, inputting the financial data into the risk identification model, and processing to obtain a risk identification result, wherein the risk identification model is used for time series analysis of the financial data to obtain time series characteristics, evaluating the time series characteristics to obtain a risk score, and generating a risk identification result according to the risk score; determining a risk processing strategy according to the risk identification result, and processing the financial account of the target user according to the risk processing strategy. Through the application, the technical problem of low recognition accuracy in identifying whether the target user handles financial business exists in the related art is solved.

Description

Technical Field

[0001] This application relates to the field of financial technology, and more specifically, to a method, apparatus, electronic device, and computer program product for handling financial risks. Background Technology

[0002] Blockchain, as a distributed technology, offers a novel solution for financial transactions and risk management. It ensures the security of transaction data through a decentralized node network. Specifically, the initiator of each transaction broadcasts the transaction details to the entire blockchain network. After verification by the nodes, the transaction information is packaged into a block and added to the distributed ledger to form a transaction record. This approach significantly improves transaction transparency and security, reducing the possibility of malicious behavior.

[0003] While blockchain technology has demonstrated significant advantages in improving transaction efficiency and enhancing data security, it still faces numerous challenges and limitations in practical applications. Firstly, current risk identification methods primarily rely on static rules and simple statistical models, which are slow to react to emerging and complex risk types, failing to detect and prevent them in a timely manner, thus missing optimal intervention opportunities. Furthermore, existing statistical models perform poorly when handling high-dimensional nonlinear data, limiting the accuracy of risk identification.

[0004] Furthermore, when faced with the massive amounts of transaction data generated in high-frequency trading scenarios, the aforementioned technologies still require manual risk management. This approach has a long response time, which reduces overall transaction security and efficiency. Traditional centralized data processing methods require uploading all user transaction data to a central server. While this approach facilitates unified management, it is prone to privacy risks.

[0005] There is currently no effective solution to the technical problem of low accuracy in identifying whether a target user is at risk when conducting financial business. Summary of the Invention

[0006] The main objective of this application is to provide a method, apparatus, electronic device, and computer program product for handling financial risks, in order to solve the technical problem of low accuracy in identifying whether a target user is at risk when conducting financial business in related technologies.

[0007] To achieve the above objectives, according to one aspect of this application, a method for handling financial risks is provided. The method includes: acquiring financial data of the target user within a preset time period, with the user's authorization; extracting a risk identification model from a blockchain, inputting the financial data into the risk identification model, and processing it to obtain a risk identification result; wherein the risk identification model is obtained by aggregating risk identification sub-models trained by M participating objects, the risk identification model is used to perform time-series analysis on the financial data to obtain time-series features, evaluating the time-series features to obtain a risk score, generating a risk identification result based on the risk score, where M is a positive integer; determining a risk handling strategy based on the risk identification result, and processing the target user's financial account according to the risk handling strategy.

[0008] Optionally, obtaining the target user's financial data within a preset time period includes: receiving a data collection instruction, collecting the target user's initial financial data from the blockchain according to the data collection instruction; obtaining cleaning rules from the blockchain, cleaning the initial financial data based on the cleaning rules to obtain cleaned financial data; obtaining a preset encryption algorithm, determining an encryption public key based on the preset encryption algorithm, and encrypting the cleaned financial data according to the encryption public key to obtain the financial data.

[0009] Optionally, the risk identification model is obtained through the following steps: obtaining M financial training samples associated with M participating objects within a historical time period from the blockchain to obtain M sets of financial training samples, wherein each set of financial training samples is processed by a preset encryption algorithm, and each set of financial training samples includes N sets of financial training data and the risk level corresponding to each set of financial training data, where N is a positive integer; using the M sets of financial training samples to train the preset risk identification model to obtain M risk identification sub-models, and aggregating the M risk identification sub-models to obtain the risk identification model.

[0010] Optionally, the process of aggregating the M risk identification sub-models to obtain the risk identification model includes: obtaining the model parameters of each risk identification sub-model to obtain M sets of model parameters, encapsulating the M sets of model parameters into a data packet, and broadcasting the data packet to Y blockchain nodes deployed in the blockchain, where Y is a positive integer; if the Y blockchain nodes receive the data packet, the Y blockchain nodes perform consensus processing on the data packet to obtain a consensus result; if the Y blockchain nodes accept the data packet without error, the step of aggregating the M risk identification sub-models to obtain the risk identification model is executed.

[0011] Optionally, the data packet is processed by Y blockchain nodes to obtain a consensus result, which includes: obtaining a consensus algorithm; each blockchain node sequentially verifies the M sets of model parameters in the data packet according to the consensus algorithm to obtain M verification results; if the verification results of K blockchain nodes indicate that any set of model parameters in the M sets of model parameters does not match the preset model rules, the Y blockchain nodes vote on the M sets of model parameters, where K is less than Y / 2 and K is a positive integer; if the vote of Y blockchain nodes passes, the data packet is confirmed to be correct.

[0012] Optionally, the M risk identification sub-models are aggregated to obtain a risk identification model in the following way: a candidate risk identification model is obtained by aggregating the M risk identification sub-models; a preset protocol is invoked to detect the candidate risk identification model based on the preset protocol; if the candidate risk identification model fails the detection, the preset risk identification model is trained using M sets of financial training samples; if the candidate risk identification model passes the detection, the candidate risk identification model is processed for consensus by Y blockchain nodes deployed in the blockchain; if the Y blockchain nodes accept the candidate risk identification model, the candidate risk identification model is determined as the risk identification model.

[0013] Optionally, before inputting the financial data into the risk identification model, the method further includes: performing homomorphic decryption on the financial data to obtain initial decrypted data; receiving a key instruction, extracting an encryption private key from the key instruction, and decrypting the initial decrypted data according to the encryption private key to obtain decrypted financial data; and performing the step of inputting the financial data into the risk identification model based on the decrypted financial data.

[0014] To achieve the above objectives, according to another aspect of this application, a financial risk processing apparatus is provided. The apparatus includes: an acquisition unit, configured to acquire financial data of a target user within a preset time period, with the user's authorization, wherein the financial data refers to business data associated with the target user's financial transactions within the preset time period; an extraction unit, configured to extract a risk identification model from a blockchain, input the financial data into the risk identification model, and process it to obtain a risk identification result, wherein the risk identification model is obtained by aggregating risk identification sub-models trained by M participating objects, the risk identification model is used to perform time-series analysis on the financial data to obtain time-series features, evaluate the time-series features to obtain a risk score, and generate a risk identification result based on the risk score, where M is a positive integer; and a determination unit, configured to determine a risk processing strategy based on the risk identification result, and process the target user's financial account according to the risk processing strategy.

[0015] According to another aspect of the present invention, a computer-readable storage medium is also provided, the computer-readable storage medium including a stored executable program, wherein, when the executable program is executed, it controls the device where the computer-readable storage medium is located to perform any of the above-mentioned financial risk processing methods.

[0016] According to another aspect of the present invention, an electronic device is also provided, including one or more processors and a memory, the memory storing an executable program, and the processor for running the program, wherein when the one or more programs are executed by the one or more processors, the one or more processors implement any of the above-described methods for handling financial risks.

[0017] According to another aspect of the present invention, a computer program product is also provided, the computer program product including a computer program, wherein the computer program, when executed by a processor, implements any of the above-described methods for handling financial risks.

[0018] In this embodiment, a financial risk handling approach is adopted. With the authorization of the target user, financial data of the target user within a preset time period is obtained. This financial data refers to business data associated with the target user's financial transactions within the preset time period. A risk identification model is extracted from the blockchain, and the financial data is input into the risk identification model to obtain a risk identification result. The risk identification model is obtained by aggregating risk identification sub-models trained by M participating objects. The risk identification model is used to perform time-series analysis on the financial data to obtain time-series features. These features are then evaluated to obtain a risk score, and a risk identification result is generated based on the risk score. M is a positive integer. A risk handling strategy is determined based on the risk identification result, and the target user's financial account is processed according to the risk handling strategy. This solves the technical problem of low accuracy in identifying whether a target user's financial transactions involve risks in related technologies. By inputting the target user's financial data into the risk identification model, processing it to obtain a risk identification result, and determining a risk handling strategy based on the risk identification result, the technical effect of improving the accuracy of identifying whether a target user's financial transactions involve risks is achieved. Attached Figure Description

[0019] The accompanying drawings, which form part of this application, are used to provide a further understanding of this application. The illustrative embodiments and descriptions of this application are used to explain this application and do not constitute an undue limitation of this application. In the drawings:

[0020] Figure 1 This is a hardware structure block diagram of a computer terminal (or mobile device) used to implement a method for handling financial risks.

[0021] Figure 2This is a flowchart of a financial risk handling method provided according to an embodiment of this application;

[0022] Figure 3 This is a schematic diagram of an optional financial risk processing system provided according to an embodiment of this application;

[0023] Figure 4 This is a schematic diagram of an optional financial risk handling method provided according to an embodiment of this application;

[0024] Figure 5 This is a schematic diagram of a data processing method provided according to an embodiment of this application;

[0025] Figure 6 This is a schematic diagram of a financial risk processing apparatus provided according to an embodiment of this application;

[0026] Figure 7 This is a structural block diagram of an electronic device according to an embodiment of this application. Detailed Implementation

[0027] To enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present application, and not all embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative effort should fall within the scope of protection of the present application.

[0028] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this application described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0029] It should be noted that all information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for display, data used for analysis, etc.) involved in this application are information and data authorized by the user or fully authorized by all parties. For example, this system has interfaces with relevant users or organizations to provide users with corresponding operation data for them to choose to agree to or refuse automated decision-making results. Before obtaining relevant information, a request for obtaining the information needs to be sent to the aforementioned user or organization through the interface, and the relevant information is obtained after receiving consent from the aforementioned user or organization; if the user chooses to refuse, the expert decision-making process is initiated. Users can view the purpose of data use in real time through authorization decoding and have the right to withdraw authorization or delete data at any time. After the authorization is withdrawn, the system will terminate the relevant data processing within 24 hours.

[0030] It should be noted that the information collected in this application is information and data authorized by the user or fully authorized by all parties, and the collection, storage, use, processing, transmission, provision, disclosure and application of the relevant data all comply with the relevant laws, regulations and standards of the relevant regions, take necessary confidentiality measures, do not violate public order and good morals, and provide corresponding operation access points for users to choose to authorize use or refuse use.

[0031] Example 1

[0032] According to an embodiment of this application, a method embodiment for handling financial risks is also provided. It should be noted that the steps shown in the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions. Furthermore, although a logical order is shown in the flowchart, in some cases, the steps shown or described may be executed in a different order than that shown here.

[0033] The method embodiment provided in Embodiment 1 of this application can be executed on a mobile terminal, computer terminal, or similar computing device. Figure 1 This is a hardware structure block diagram of a computer terminal (or mobile device) used to implement a method for handling financial risks, such as... Figure 1 As shown, computer terminal 10 (or mobile device) may include one or more ( Figure 1The processor 102 (which may include, but is not limited to, a microprocessor MCU (Microcontroller Unit) or a programmable gate array (FPGA)) is shown as 102a, 102b, ..., 102n. It also includes a memory 104 for storing data and a transmission device 106 for communication functions. In addition, it may include: a display, an input / output interface, a Universal Serial Bus (USB) port (which may be included as one of the ports of a BUS bus), a network interface, a keyboard, a cursor control device, a power supply, and / or a camera. Those skilled in the art will understand that... Figure 1 The structure shown is for illustrative purposes only and does not limit the structure of the aforementioned electronic device. For example, computer terminal 10 may also include... Figure 1 The more or fewer components shown, or having the same Figure 1 The different configurations shown.

[0034] It should be noted that the aforementioned one or more processors 102 and / or other data processing circuits are generally referred to herein as "data processing circuits". These data processing circuits may be embodied, in whole or in part, in software, hardware, firmware, or any other combination thereof. Furthermore, the data processing circuits may be a single, independent processing module, or may be integrated, in whole or in part, into any other element within the computer terminal 10 (or mobile device). As involved in the embodiments of this application, the data processing circuits serve as a processor control mechanism (e.g., selection of a variable resistor termination path connected to an interface).

[0035] The memory 104 can be used to store software programs and modules of application software, such as the program instructions / data storage device corresponding to the financial risk processing method in this embodiment. The processor 102 executes various functional applications and data processing by running the software programs and modules stored in the memory 104, thereby realizing the aforementioned financial risk processing method. The memory 104 may include high-speed random access memory, and may also include non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some instances, the memory 104 may further include memory remotely located relative to the processor 102, and these remote memories can be connected to the computer terminal 10 via a network. Examples of such networks include, but are not limited to, the Internet, corporate intranets, local area networks, mobile communication networks, and combinations thereof.

[0036] The transmission device 106 is used to receive or send data via a network. Specific examples of the network described above may include a wireless network provided by the communication provider of the computer terminal 10. In one example, the transmission device 106 includes a network interface controller (NIC) and a network interface, which can be connected to other network devices via a base station to communicate with the Internet. In another example, the transmission device 106 may be a radio frequency (RF) module, used for wireless communication with the Internet.

[0037] The display can be, for example, a touchscreen liquid crystal display (LCD), which allows the user to interact with the user interface of the computer terminal 10 (or mobile device).

[0038] Under the aforementioned operating environment, this application provides the following: Figure 2 The methods for handling financial risks are shown. Figure 2 This is a flowchart of a financial risk handling method provided according to an embodiment of this application, such as... Figure 2 As shown, the method includes the following steps:

[0039] Step S201: With the authorization of the target user, obtain the target user's financial data within a preset time period, wherein the financial data refers to the business data associated with the target user when conducting financial transactions within the preset time period.

[0040] Specifically, to improve the security of financial transactions, it is necessary to obtain the financial data of a target user before identifying whether there are any risks associated with the financial transactions they have conducted. It should be noted that explicit authorization from the target user is required before obtaining this data. After the target user consents to the data acquisition, financial data associated with the target user's financial transactions can be obtained from multiple data sources. This financial data can include various aspects of the transaction, such as account statements from financial institutions and transaction frequency; the financial transaction can refer to the transactions and service activities between the user and the financial institution, including deposits and withdrawals.

[0041] Step S202: Extract the risk identification model from the blockchain, input the financial data into the risk identification model, and process it to obtain the risk identification result. The risk identification model is obtained by aggregating the risk identification sub-models trained by M participating objects. The risk identification model is used to perform time series analysis on the financial data to obtain time series features, evaluate the time series features to obtain a risk score, and generate the risk identification result based on the risk score. M is a positive integer.

[0042] Specifically, after obtaining financial data, a risk identification model can be extracted from the blockchain to conduct risk analysis. The target user's financial data is then input into this model to process and obtain the risk identification results for financial transactions. This risk identification model can be formed by aggregating risk identification sub-models trained by multiple participants, including financial institutions and enterprises, through the blockchain's parameter aggregation mechanism. This model can perform time-series analysis on the financial data, analyzing its trends over time and extracting temporal features. Then, a Bayesian network is used to dynamically calculate the risk score based on these temporal features. Finally, a risk identification result indicating the risk level of the transaction is generated based on the risk score; a higher score indicates a greater transaction risk. For example, if the risk score exceeds a pre-set threshold, the transaction can be labeled as "high-risk," and actions such as freezing the account and requiring more verification information can be taken.

[0043] Step S203: Determine the risk handling strategy based on the risk identification results, and process the target user's financial account according to the risk handling strategy.

[0044] Specifically, after obtaining the risk identification results, the risk level can be determined based on the results, and then the corresponding risk handling strategy can be determined based on the risk level. The risk handling strategy may include suspending or canceling transactions, triggering account freezing operations, etc.

[0045] It should be noted that the target user's financial account refers to the location where financial assets, such as accounts at financial institutions undergoing risk assessment, are held. Once a specific risk management strategy is obtained, specific actions can be taken on the financial account, such as restricting account functions, freezing the account, or even resolving the risk, thereby protecting the security of both the financial institution and the user.

[0046] The financial risk handling method provided in this application embodiment obtains the financial data of the target user within a preset time period with the user's authorization. This financial data refers to the business data associated with the target user's financial transactions within the preset time period. A risk identification model is extracted from the blockchain, and the financial data is input into the risk identification model to obtain a risk identification result. The risk identification model is obtained by aggregating risk identification sub-models trained by M participating objects. The risk identification model is used to perform time-series analysis on the financial data to obtain time-series features, evaluate these features to obtain a risk score, and generate a risk identification result based on the risk score (M is a positive integer). A risk handling strategy is determined based on the risk identification result, and the target user's financial account is processed according to the risk handling strategy. This method solves the technical problem of low accuracy in identifying whether a target user's financial transactions involve risks in related technologies. By inputting the target user's financial data into the risk identification model, processing it to obtain a risk identification result, and determining a risk handling strategy based on the risk identification result, the method achieves the technical effect of improving the accuracy of identifying whether a target user's financial transactions involve risks.

[0047] Optionally, in the financial risk handling method provided in this application embodiment, obtaining the financial data of the target user within a preset time period includes: receiving a data collection instruction, collecting the initial financial data of the target user from the blockchain according to the data collection instruction; obtaining cleaning rules from the blockchain, cleaning the initial financial data based on the cleaning rules to obtain cleaned financial data; obtaining a preset encryption algorithm, determining an encryption public key based on the preset encryption algorithm, and encrypting the cleaned financial data according to the encryption public key to obtain financial data.

[0048] Specifically, after authorization from the target user, in order to obtain financial data, the initial financial data of the target user can be collected through the API (Application Programming Interface) interface in the blockchain network after receiving the data collection instruction. The data collection instruction can include information such as the specified target user, the time range for data collection, and the data type. It should be noted that when obtaining the initial financial data, interaction with the underlying blockchain can be performed to retrieve the target user's financial transaction records within a preset time period from the distributed ledger, thus obtaining the initial financial data.

[0049] Furthermore, after obtaining the initial financial data, cleaning rules can be retrieved from the blockchain's smart contracts to avoid subjective biases or errors in the data processing. Then, based on these cleaning rules, the initial financial data is cleaned, including steps such as removing duplicate data, correcting formatting errors, filling missing values, and normalizing values, resulting in cleaned financial data. For example, the cleaning rule could be to remove duplicate transaction records within one minute before and after all transaction timestamps. The cleaned financial data is then encrypted, which can be done using a preset encryption algorithm (such as isomorphic encryption). This involves determining the encryption public key based on the current encryption algorithm and then using this public key to encrypt the cleaned financial data, ensuring data security during transmission and storage, thus obtaining the final financial data.

[0050] This embodiment processes initial financial data to obtain financial data, which not only ensures that the collected and processed financial data is accurate, complete, and free from interference, but also allows the data to be securely used for risk identification while protecting data privacy.

[0051] Optionally, in the financial risk processing method provided in this application embodiment, the risk identification model is obtained through the following steps: obtaining M financial training samples associated with M participating objects within a historical time period from the blockchain to obtain M sets of financial training samples, wherein each set of financial training samples is obtained by processing through a preset encryption algorithm, and each set of financial training samples includes N sets of financial training data and the risk level corresponding to each set of financial training data, where N is a positive integer; training the preset risk identification model using the M sets of financial training samples to obtain M risk identification sub-models, and aggregating the M risk identification sub-models to obtain the risk identification model.

[0052] To utilize the risk identification model for risk assessment, the model first needs to be trained. Specifically, firstly, financial training samples related to multiple participants over a historical period are obtained from the blockchain and processed using a pre-defined encryption algorithm. Then, these training samples are used to train a sub-model for each participant. Each set of financial training samples contains feature information related to financial transactions (i.e., financial training data) and a risk level. After obtaining the encrypted financial training samples from the blockchain, the pre-defined risk identification model is trained using each set of financial training samples, thus obtaining the corresponding risk identification sub-model. It should be noted that the training process must maintain data encryption and use homomorphic encryption technology so that the model can learn the risk patterns in the data without actually decrypting the data, thereby optimizing its parameters and improving its risk identification capabilities.

[0053] Furthermore, after the risk identification sub-model associated with each participating object is trained, these sub-models can be aggregated to build a more powerful and comprehensive risk identification model. The aggregation process can be achieved through federated learning, which can leverage the data advantages of different participating objects to comprehensively generate a model that can more broadly cover various financial risk scenarios. For example, weighted averaging or parameter fusion algorithms can be used for the sub-model parameters to make the generated risk identification model not only highly accurate but also have strong generalization ability and can adapt to the ever-changing financial environment.

[0054] This embodiment uses encrypted financial data for distributed model training and then aggregates these training results to obtain a risk identification model. This effectively resolves the contradiction between financial data privacy protection and model training, enhances data diversity and model robustness, and achieves secure data sharing and high-performance model training.

[0055] Optionally, in the financial risk processing method provided in this application embodiment, the step of aggregating M risk identification sub-models to obtain a risk identification model includes: obtaining model parameters of each risk identification sub-model to obtain M sets of model parameters, encapsulating the M sets of model parameters into a data packet, and broadcasting the data packet to Y blockchain nodes deployed in the blockchain, where Y is a positive integer; if the Y blockchain nodes receive the data packet, the Y blockchain nodes perform consensus processing on the data packet to obtain a consensus result; if the Y blockchain nodes accept the data packet without error, the step of aggregating the M risk identification sub-models to obtain a risk identification model is executed.

[0056] Specifically, after obtaining the risk identification sub-model for each participant, to aggregate these models, it is first necessary to extract model parameters from their risk identification sub-models, forming multiple sets of model parameters. These model parameters may include weight matrices, bias terms, and other key values ​​obtained during training. Then, the model parameters are encapsulated into data packets. For example, a preset encryption standard (such as RSA encryption (Rivest-Shamir-Adleman, RSA public-key encryption, an asymmetric encryption algorithm)) can be used to encrypt the model parameters, adding the identity identifiers and digital signatures of different participants. This data is then combined into a single data packet, transforming the model parameters into a format suitable for transmission over the network, facilitating broadcasting of the model parameters within the blockchain network while protecting them from tampering. The data packet is then broadcast to multiple blockchain nodes deployed in the blockchain, where these nodes are responsible for verifying the data packet and processing consensus.

[0057] Furthermore, after each blockchain node receives the data packet, they perform consensus processing to obtain a consensus result. This means that the blockchain nodes verify the compliance and integrity of the data packet according to a pre-defined consensus mechanism. At this point, each node can use cryptographic methods to verify the digital signature in the data packet, confirm the sender's identity, and verify whether the encapsulated parameters conform to the expected format and rules. For example, they can check whether the parameters conflict with previous versions or exceed a preset numerical range. Consensus processing means that only when a majority of nodes verify and agree on the model parameters can these parameters be included as part of the aggregation process. Once consensus processing is complete and all nodes consider the data packet correct, the aggregation operation of the sub-model can continue. At this point, the parameters of the risk identification sub-model can be weighted and averaged, or other federated learning algorithms can be used to synthesize the inputs of all participating nodes, generating a more comprehensive risk identification model.

[0058] This embodiment aggregates multiple risk identification sub-models to obtain a risk identification model, eliminating the "black box" effect of model training, protecting the data privacy of financial institutions, optimizing the efficiency of parameter aggregation, reducing unnecessary data transmission and processing delays, avoiding model bias caused by data quality issues, and improving the overall performance of the risk identification model.

[0059] Optionally, in the financial risk handling method provided in this application embodiment, the consensus processing of the data packet by Y blockchain nodes to obtain the consensus result includes: obtaining a consensus algorithm; each blockchain node sequentially verifying M sets of model parameters in the data packet according to the consensus algorithm to obtain M verification results; if the verification results of K blockchain nodes indicate that any set of model parameters in the M sets of model parameters does not match the preset model rules, the Y blockchain nodes vote on the M sets of model parameters, where K is less than Y / 2 and K is a positive integer; if the vote of Y blockchain nodes passes, the data packet is confirmed to be correct.

[0060] Specifically, when blockchain nodes perform consensus processing, they first obtain a consensus algorithm suitable for the current context. This algorithm is used to verify the model parameters in the data packet. Then, each blockchain node sequentially verifies the model parameters in the data packet according to the consensus algorithm. For example, it checks whether the model parameters meet preset model rules, including parameter format, numerical range, and compatibility with the model architecture. After each node completes its verification, it generates a corresponding verification result, indicating whether the node approves the set of parameters.

[0061] During the verification process, if less than half of the total number of nodes finds that a certain set of model parameters does not conform to the preset model rules, meaning that the verification results of less than half of the total number of nodes indicate that any set of model parameters does not match the preset model rules, a voting process needs to be initiated. This involves all blockchain nodes participating in a vote to determine whether this set of model parameters should be adopted. The voting mechanism enables "democratic decision-making." Even if some nodes raise objections, as long as a majority of nodes (i.e., more than half) believe the parameters are valid, these parameters can continue to be used for model aggregation. When a majority of the votes from the blockchain nodes passes, it can be confirmed that the model parameters in the data packet are correct and can be used in subsequent model aggregation steps.

[0062] This embodiment utilizes blockchain and consensus mechanisms to establish a model parameter verification and confirmation process, effectively screening out high-quality parameters, avoiding the impact of inefficient or harmful parameters on model performance, greatly improving security protection capabilities, and promoting fairness and transparency in data processing.

[0063] Optionally, in the financial risk processing method provided in this application embodiment, M risk identification sub-models are aggregated to obtain a risk identification model in the following manner: M risk identification sub-models are aggregated to obtain a candidate risk identification model; a preset protocol is invoked, and the candidate risk identification model is tested based on the preset protocol; if the candidate risk identification model fails the test, the preset risk identification model is trained using M sets of financial training samples; if the candidate risk identification model passes the test, Y blockchain nodes deployed in the blockchain perform consensus processing on the candidate risk identification model; if the Y blockchain nodes accept the candidate risk identification model, the candidate risk identification model is determined as the risk identification model.

[0064] Specifically, after the consensus operation of the blockchain nodes is completed, the parameters of all risk identification sub-models can be processed by weighted averaging, model fusion, or parameter transfer to aggregate candidate risk identification models. Then, a preset protocol (i.e., a smart contract) is invoked to test the candidate risk identification models based on the preset protocol, such as checking whether the model performance indicators meet the standards and whether the prediction accuracy of the model reaches a predetermined threshold.

[0065] If a candidate risk identification model fails the detection, it indicates that the model does not meet expectations in some aspects. In this case, the preset risk identification model needs to be retrained using multiple sets of financial training samples until it passes the detection of the preset protocol to achieve better performance. Conversely, if the candidate risk identification model passes the detection, the blockchain nodes deployed in the blockchain need to perform consensus processing on the candidate risk identification model. That is, the blockchain nodes jointly verify the validity and security of the candidate risk identification model by executing a consensus algorithm to ensure that it meets the requirements of network consensus. Consensus processing can prevent the possibility of malicious tampering of the model. Only when a majority of the nodes in all blockchain nodes accept the candidate risk identification model can it be determined as a risk identification model.

[0066] This embodiment combines federated learning and blockchain technology, integrating parameters from multiple risk identification sub-models to enable the risk identification model to comprehensively consider the characteristics of transaction data from different entities, thereby improving model quality and helping financial institutions enhance their risk identification capabilities.

[0067] Optionally, in the financial risk processing method provided in this application embodiment, before inputting the financial data into the risk identification model, the method further includes: performing homomorphic decryption operation on the financial data to obtain initial decrypted data; receiving a key instruction, extracting an encryption private key from the key instruction, and decrypting the initial decrypted data according to the encryption private key to obtain decrypted financial data; and performing the step of inputting the financial data into the risk identification model based on the decrypted financial data.

[0068] Specifically, after obtaining the target user's data, since the financial data is encrypted during transmission and storage, it is necessary to perform homomorphic decryption on the financial data before using it for risk identification. This process maintains the computational capability of the encrypted data and yields the initial decrypted data. Homomorphic decryption is the reverse operation of homomorphic encryption, which allows for necessary calculations while maintaining the encrypted state of the data, so that the subsequent decryption process can proceed smoothly.

[0069] Once the initial decryption data is ready, a key instruction containing the encrypted private key paired with the public key can be received. The encrypted private key is then extracted and used to decrypt the initial decryption data, restoring it to a understandable and analyzable form—the decrypted financial data. Finally, the decrypted financial data is input into a risk identification model, which then performs a risk assessment.

[0070] This embodiment decrypts financial data, protecting its privacy while effectively utilizing it for risk identification model assessment. This not only enhances data security but also improves the efficiency of transaction risk management, avoiding privacy leaks and security threats encountered in traditional data decryption processes.

[0071] This application also provides a method for handling financial risks. Figure 3 This is a schematic diagram of an optional financial risk processing system provided according to an embodiment of this application, such as... Figure 3 As shown, the system includes: a business application layer, a tool and interface layer, a component layer, and an underlying management and control layer for connecting ecosystem roles and requirements.

[0072] The business application layer is associated with participating entities (such as financial institutions, enterprises, and data providers), risk supervision entities (such as regulatory agencies), and service entities. Participating entities, as the "demand initiators" of the system, can trigger risk handling processes. Risk supervision entities can achieve penetrating supervision of financial activities through blockchain-stored model logs and transaction records. Service entities submit model parameters of the risk identification model to the system through the model training interface, thereby scheduling the distributed computing power of the blockchain to complete on-chain training, and calling the trained risk identification model to output risk identification results. The model's compliance and accuracy can also be verified through blockchain-stored model versions and training process logs. The tools and interface layer includes a model training interface, a software development kit (SDK), and an application programming interface (API). The model training interface is used for on-chain training of the risk identification model and can realize functions such as uploading model parameters, scheduling training tasks (e.g., allocating them to multiple blockchain nodes for parallel computing), and monitoring training progress. The API can integrate various services and provide a standardized call entry point for upper-layer business applications. For example, risk supervision objects can obtain the risk identification results of the risk identification model through this interface. The SSD provides tools for blockchain development (e.g., smart contract writing), model calling, and risk data processing (e.g., abnormal transaction filtering), reducing the difficulty of system integration.

[0073] The component layer includes a strategy execution module, a risk management module, a blockchain evidence storage module, and an on-chain management module. The strategy execution module's functions include risk handling strategy development, smart contract triggering, and execution. The risk management module's functions include real-time monitoring, on-chain assessment, decision-making, and handling. For example, it can collect financial data, output risk identification results through a risk identification model, determine risk handling strategies based on these results, and finally trigger actions on the target user's financial account (such as freezing the account or adjusting credit limits) according to the risk handling strategy. The blockchain evidence storage module's functions include model version evidence storage and transaction record tracing. The on-chain management module's functions include on-chain training, version updates, and access control. On-chain training refers to using the computing power of distributed blockchain nodes, employing federated learning and homomorphic encryption technologies to complete model training while protecting user data privacy. Version updates refer to the requirement that after model iteration, the new version must undergo consensus verification by multiple blockchain nodes before it can be deployed. Access control means that only service recipients can modify model parameters using their private keys; other roles (such as participating entities) can only access the risk identification results output by the model. The underlying control layer provides data storage, computing power support, and security for the upper layer. It can include a data control platform, blockchain underlying services, a distributed database, and a model engine. The data control platform stores the financial data of the target users. The blockchain underlying services provide basic functions such as transactions, permissions, and encryption. The distributed database stores the core data of the target users (such as user identity information and transaction history) and model training data. It uses sharding technology to improve read and write efficiency and is linked with the blockchain ledger. The model engine is used to manage the risk identification model on-chain throughout its entire lifecycle, including: on-chain training (calling model training data in the distributed database and completing model training through federated learning), inference (calling the trained risk identification model to output risk identification results), and evidence storage (storing the model version, training logs, and risk identification results in the blockchain).

[0074] The aforementioned system is capable of implementing an optional financial risk management method. Figure 4 This is a schematic diagram of an optional financial risk handling method provided according to an embodiment of this application, such as... Figure 4 As shown, the method includes:

[0075] First, a blockchain supporting on-chain training and privacy-preserving computation of the risk identification model can be constructed. Then, a distributed environment is set up for the on-chain training of the risk identification model, and the execution rules for privacy computation are configured, i.e., initialization configuration is performed. When risk detection is needed for the financial transactions of a target user, the target user's financial data can be collected based on blockchain nodes and privacy computation technology. This data is then input into the risk identification model, which performs on-chain preprocessing and generates risk identification results. Specifically, the risk identification model can perform time-series analysis on the financial data to obtain time-series features, evaluate these features to obtain a risk score, and generate risk identification results based on the risk score.

[0076] Furthermore, after obtaining the risk identification results, to generate the optimal risk management strategy, multiple blockchain nodes deployed on the blockchain can engage in consensus processing and voting. The risk management strategy is then generated if the blockchain nodes approve the vote. Once the specific risk management strategy is obtained, specific operations can be implemented on the financial account, such as restricting account functions, adjusting credit limits, or temporarily freezing the account until the risk is resolved, thereby protecting the assets of financial institutions and users. It should be noted that during the above operations, blockchain technology can be used for on-chain strategy execution and full-process notarization, making financial transactions completely transparent. After implementing specific operations on the financial account, the risk identification model deployed on the blockchain can be used to identify risks in real-time synchronized transaction data, i.e., to perform post-transaction risk detection and dynamic control, thereby improving transaction security.

[0077] It should be noted that in order to use the risk identification model for risk assessment, the model first needs to be trained. Specifically, firstly, financial training samples related to multiple participants over a historical period are obtained from the blockchain and processed using a preset encryption algorithm. Then, these training samples are used to train a sub-model for each participant, thereby obtaining a corresponding risk identification sub-model. Finally, these sub-models are aggregated to obtain the risk identification model. Figure 5 This is a schematic diagram of a data processing method provided according to an embodiment of this application, such as... Figure 5 As shown, to obtain financial training samples, data collection can first be performed to gather financial transaction data and risk data related to different participants. Next, this data is cleaned. Cleaning rules can be obtained from the blockchain, and the data is cleaned based on these rules to obtain the cleaned data.

[0078] Furthermore, a preset encryption algorithm is obtained, a public key is determined based on homomorphic encryption, and the cleaned data is encrypted according to the public key to obtain financial training samples corresponding to each participant. These samples are then transmitted to the model processing node in the blockchain. The node performs federated learning training based on multiple financial training samples. In other words, the model processing node trains its own risk identification sub-model based on different financial training samples. The risk identification sub-models are then aggregated into a global model to obtain the risk identification model.

[0079] It's important to note that during training, the distributed ledger and smart contract features of blockchain can be leveraged to achieve data availability without visibility during model training. Through encrypted parameter exchange and on-chain verification mechanisms, the original data is always kept locally to protect privacy, while simultaneously providing a trusted foundation for on-chain training. The entire model parameter aggregation process is fully recorded on the blockchain, ensuring the traceability of the training process. Furthermore, the model parameters and training logs of the trained risk identification model are stored on the blockchain, enabling traceability of model version updates.

[0080] It should be noted that after obtaining the trained risk identification model, the model can be verified on-chain. For example, a smart contract can be used to check whether the model's performance metrics meet the standards. If they do, the model version is updated; if they do not, a retraining process is triggered. The model that has passed on-chain verification is then further trained or fine-tuned, and the adjusted risk identification model is stored on the blockchain for deployment.

[0081] This embodiment inputs the financial data of the target user into a risk identification model, processes it to obtain risk identification results, and determines risk handling strategies based on the risk identification results, thereby achieving the technical effect of improving the accuracy of identifying whether there are risks in the target user's financial business transactions.

[0082] It should be noted that the steps shown in the flowchart in the accompanying drawings can be executed in a computer such as a set of computer-executable instructions, and although a logical order is shown in the flowchart, in some cases the steps shown or described may be executed in a different order than that shown here.

[0083] Example 2

[0084] This application also provides a financial risk processing apparatus. It should be noted that the financial risk processing apparatus of this application can be used to execute the financial risk processing method provided in this application. The following describes the financial risk processing apparatus provided in this application.

[0085] According to an embodiment of this application, an apparatus for implementing the above-described method for handling financial risks is also provided. Figure 6This is a schematic diagram of a financial risk processing device provided according to an embodiment of this application, such as... Figure 6 As shown, the device includes: an acquisition unit 60, an extraction unit 61, and a determination unit 62.

[0086] The acquisition unit 60 is used to acquire the financial data of the target user within a preset time period with the authorization of the target user. The financial data refers to the business data associated with the target user when conducting financial business within the preset time period.

[0087] Extraction unit 61 is used to extract the risk identification model from the blockchain, input financial data into the risk identification model, process it to obtain the risk identification result, wherein the risk identification model is obtained by aggregating the risk identification sub-models trained by M participating objects, the risk identification model is used to perform time series analysis on financial data to obtain time series features, evaluate the time series features to obtain a risk score, and generate the risk identification result based on the risk score, where M is a positive integer;

[0088] The determining unit 62 is used to determine the risk handling strategy based on the risk identification results, and to process the financial account of the target user according to the risk handling strategy.

[0089] The financial risk processing apparatus provided in this application embodiment acquires the financial data of the target user within a preset time period with the authorization of the target user by the acquisition unit 60. The financial data refers to the business data associated with the target user's financial transactions within the preset time period. The extraction unit 61 extracts a risk identification model from the blockchain, inputs the financial data into the risk identification model, and processes it to obtain a risk identification result. The risk identification model is obtained by aggregating risk identification sub-models trained by M participating objects. The risk identification model is used to perform time-series analysis on the financial data to obtain time-series features, evaluates the time-series features to obtain a risk score, and generates a risk identification result based on the risk score, where M is a positive integer. The determination unit 62 determines a risk processing strategy based on the risk identification result and processes the target user's financial account according to the risk processing strategy. This solves the technical problem of low accuracy in identifying whether a target user's financial transactions involve risks in related technologies. By inputting the target user's financial data into the risk identification model, processing it to obtain a risk identification result, and determining a risk processing strategy based on the risk identification result, the technical effect of improving the accuracy of identifying whether a target user's financial transactions involve risks is achieved.

[0090] Optionally, in the financial risk processing apparatus provided in this application embodiment, the acquisition unit 60 includes: a receiving module, used to receive a data acquisition instruction and acquire initial financial data of a target user from the blockchain according to the data acquisition instruction; a first acquisition module, used to acquire cleaning rules from the blockchain, perform data cleaning on the initial financial data based on the cleaning rules, and obtain cleaned financial data; and a second acquisition module, used to acquire a preset encryption algorithm, determine an encryption public key based on the preset encryption algorithm, and encrypt the cleaned financial data according to the encryption public key to obtain financial data.

[0091] Optionally, in the financial risk processing apparatus provided in this application embodiment, the extraction unit 61 includes: a third acquisition module, used to acquire M financial training samples associated with M participating objects within a historical time period from the blockchain, to obtain M sets of financial training samples, wherein each set of financial training samples is obtained by processing through a preset encryption algorithm, and each set of financial training samples includes N sets of financial training data and the risk level corresponding to each set of financial training data, where N is a positive integer; and a training module, used to train a preset risk identification model using the M sets of financial training samples to obtain M risk identification sub-models, and to aggregate the M risk identification sub-models to obtain a risk identification model.

[0092] Optionally, in the financial risk processing apparatus provided in this application embodiment, the extraction unit 61 includes: a fourth acquisition module, used to acquire model parameters of each risk identification sub-model, obtain M sets of model parameters, encapsulate the M sets of model parameters into a data packet, and broadcast the data packet to Y blockchain nodes deployed in the blockchain, where Y is a positive integer; a first processing module, used to perform consensus processing on the data packet by the Y blockchain nodes when the data packet is received, and obtain a consensus result; and an execution module, used to execute the step of aggregating the M risk identification sub-models to obtain a risk identification model when the data packet is received correctly by the Y blockchain nodes.

[0093] Optionally, in the financial risk processing apparatus provided in this application embodiment, the extraction unit 61 includes: a fifth acquisition module, used to acquire a consensus algorithm, wherein each blockchain node sequentially verifies M sets of model parameters in the data packet according to the consensus algorithm to obtain M verification results; a voting module, used to have Y blockchain nodes vote on the M sets of model parameters if the verification results of K blockchain nodes indicate that any set of model parameters in the M sets of model parameters does not match the preset model rules, wherein K is less than Y / 2 and K is a positive integer; and a confirmation module, used to confirm that the data packet is correct if the votes of Y blockchain nodes pass.

[0094] Optionally, in the financial risk processing apparatus provided in this application embodiment, the extraction unit 61 includes: an aggregation module, used to aggregate M risk identification sub-models to obtain a candidate risk identification model; a calling module, used to call a preset protocol, detect the candidate risk identification model based on the preset protocol, and if the candidate risk identification model fails the detection, perform the step of training the preset risk identification model using M sets of financial training samples; and a second processing module, used to, if the candidate risk identification model passes the detection, have Y blockchain nodes deployed in the blockchain perform consensus processing on the candidate risk identification model, and if the Y blockchain nodes accept the candidate risk identification model, determine the candidate risk identification model as a risk identification model.

[0095] Optionally, in the financial risk processing apparatus provided in this application embodiment, the apparatus further includes: a processing unit, configured to perform homomorphic decryption processing on the financial data before inputting the financial data into the risk identification model to obtain initial decrypted data; a receiving unit, configured to receive a key instruction, extract an encryption private key from the key instruction, and decrypt the initial decrypted data according to the encryption private key to obtain decrypted financial data; and an execution unit, configured to execute the step of inputting the financial data into the risk identification model based on the decrypted financial data.

[0096] It should be noted that the acquisition unit 60, extraction unit 61, and determination unit 62 mentioned above correspond to steps S201 to S203 in Embodiment 1. The instances and application scenarios implemented by the above units and corresponding steps are the same, but are not limited to the content disclosed in Embodiment 1. It should be noted that the above modules or units can be hardware or software components stored in memory (e.g., memory 104) and processed by one or more processors (e.g., processors 102a, 102b, ..., 102n). The above units can also be part of a device and can run in the computer terminal 10 provided in Embodiment 1.

[0097] Example 3

[0098] Embodiments of this application may provide a computer terminal, which may be any computer terminal device in a group of computer terminals. Optionally, in this embodiment, the aforementioned computer terminal may also be replaced with a mobile terminal or an electronic device, etc.

[0099] Optionally, in this embodiment, the computer terminal may be located in at least one of a plurality of network devices in a computer network.

[0100] In this embodiment, the computer terminal described above can execute the program code for the following steps in the financial risk handling method: With the authorization of the target user, acquire the target user's financial data within a preset time period, wherein the financial data refers to the business data associated with the target user's financial transactions within the preset time period; extract a risk identification model from the blockchain, input the financial data into the risk identification model, and process it to obtain a risk identification result, wherein the risk identification model is obtained by aggregating risk identification sub-models trained by M participating objects, the risk identification model is used to perform time-series analysis on the financial data to obtain time-series features, evaluate the time-series features to obtain a risk score, generate a risk identification result based on the risk score, where M is a positive integer; determine a risk handling strategy based on the risk identification result, and process the target user's financial account according to the risk handling strategy.

[0101] Optionally, the aforementioned computer terminal may execute the program code for the following steps in the financial risk handling method: receiving a data collection instruction, collecting the target user's initial financial data from the blockchain according to the data collection instruction; obtaining cleaning rules from the blockchain, cleaning the initial financial data based on the cleaning rules to obtain cleaned financial data; obtaining a preset encryption algorithm, determining an encryption public key based on the preset encryption algorithm, and encrypting the cleaned financial data according to the encryption public key to obtain financial data.

[0102] Optionally, the aforementioned computer terminal can execute the program code for the following steps in the financial risk handling method: obtaining M financial training samples associated with M participating objects within a historical time period from the blockchain to obtain M sets of financial training samples, wherein each set of financial training samples is obtained by processing through a preset encryption algorithm, and each set of financial training samples includes N sets of financial training data and the risk level corresponding to each set of financial training data, where N is a positive integer; using the M sets of financial training samples to train a preset risk identification model to obtain M risk identification sub-models, and aggregating the M risk identification sub-models to obtain a risk identification model.

[0103] Optionally, the aforementioned computer terminal can execute the following steps in the financial risk processing method: obtain the model parameters of each risk identification sub-model, obtain M sets of model parameters, encapsulate the M sets of model parameters into a data packet, and broadcast the data packet to Y blockchain nodes deployed in the blockchain, where Y is a positive integer; if the Y blockchain nodes receive the data packet, the Y blockchain nodes perform consensus processing on the data packet to obtain a consensus result; if the Y blockchain nodes accept the data packet without error, execute the step of aggregating the M risk identification sub-models to obtain a risk identification model.

[0104] Optionally, the aforementioned computer terminal can execute the program code for the following steps in the financial risk handling method: obtaining a consensus algorithm; each blockchain node sequentially verifies the M sets of model parameters in the data packet according to the consensus algorithm, obtaining M verification results; if the verification results of K blockchain nodes indicate that any set of model parameters in the M sets of model parameters does not match the preset model rules, Y blockchain nodes vote on the M sets of model parameters, where K is less than Y / 2 and K is a positive integer; if the votes of Y blockchain nodes pass, the data packet is confirmed to be correct.

[0105] Optionally, the aforementioned computer terminal can execute the program code for the following steps in the financial risk processing method: aggregating M risk identification sub-models to obtain a candidate risk identification model; invoking a preset protocol to detect the candidate risk identification model based on the preset protocol; if the candidate risk identification model fails the detection, executing the step of training the preset risk identification model using M sets of financial training samples; if the candidate risk identification model passes the detection, Y blockchain nodes deployed in the blockchain perform consensus processing on the candidate risk identification model; if the Y blockchain nodes accept the candidate risk identification model, the candidate risk identification model is determined as the risk identification model.

[0106] Optionally, the aforementioned computer terminal can execute program code for the following steps in the financial risk processing method: performing homomorphic decryption on the financial data to obtain initial decrypted data; receiving a key instruction, extracting the encryption private key from the key instruction, and decrypting the initial decrypted data according to the encryption private key to obtain decrypted financial data; and executing the step of inputting the financial data into the risk identification model based on the decrypted financial data.

[0107] Optionally, Figure 7 This is a structural block diagram of an electronic device according to an embodiment of this application. Figure 7 As shown, the electronic device may include: one or more ( Figure 7 (Only one is shown) processor 702, memory 704, memory controller, and peripheral interface, wherein the peripheral interface is connected to the radio frequency module, audio module and display.

[0108] The memory can be used to store software programs and modules, such as the program instructions / modules corresponding to the financial risk processing method and apparatus in this application embodiment. The processor executes various functional applications and data processing by running the software programs and modules stored in the memory, thereby realizing the aforementioned financial risk processing method. The memory may include high-speed random access memory, and may also include non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some instances, the memory may further include memory remotely located relative to the processor, and these remote memories can be connected to the terminal via a network. Examples of such networks include, but are not limited to, the Internet, corporate intranets, local area networks, mobile communication networks, and combinations thereof.

[0109] The processor can access the information and application programs stored in the memory via the transmission device to execute the steps described above in the financial risk handling method.

[0110] Those skilled in the art will understand that Figure 7 The structure shown is for illustrative purposes only. Electronic devices can also be smartphones, tablets, handheld computers, mobile internet devices (MIDs), PADs, and other terminal devices. Figure 7 This does not limit the structure of the aforementioned electronic device. For example, electronic devices may also include components that are more... Figure 7 The more or fewer components shown (such as network interfaces, display devices, etc.), or having the same Figure 7 The different configurations shown.

[0111] Those skilled in the art will understand that all or part of the steps in the various methods of the above embodiments can be implemented by a program instructing the hardware related to the terminal device. The program can be stored in a computer-readable storage medium, which may include: flash drive, read-only memory (ROM), random access memory (RAM), disk or optical disk, etc.

[0112] Example 4

[0113] Embodiments of this application also provide a storage medium. Optionally, in this embodiment, the storage medium can be used to store the program code executed by the financial risk handling method provided in Embodiment 1.

[0114] Optionally, in this embodiment, the storage medium may be located in any computer terminal in a group of computer terminals in a computer network, or in any mobile terminal in a group of mobile terminals.

[0115] Optionally, in this embodiment, the storage medium is configured to store program code for performing the following steps: With the authorization of the target user, acquire the target user's financial data within a preset time period, wherein the financial data refers to the business data associated with the target user's financial transactions within the preset time period; extract a risk identification model from the blockchain, input the financial data into the risk identification model, and process it to obtain a risk identification result, wherein the risk identification model is obtained by aggregating risk identification sub-models trained by M participating objects, the risk identification model is used to perform time-series analysis on the financial data to obtain time-series features, evaluate the time-series features to obtain a risk score, generate a risk identification result based on the risk score, where M is a positive integer; determine a risk handling strategy based on the risk identification result, and process the target user's financial account according to the risk handling strategy.

[0116] This application also provides a computer program product that, when executed on a data processing device, is suitable for performing steps of a financial risk processing method.

[0117] The sequence numbers of the embodiments in this application are for descriptive purposes only and do not represent the superiority or inferiority of the embodiments.

[0118] In the above embodiments of this application, the descriptions of each embodiment have different focuses. For parts not described in detail in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0119] In the several embodiments provided in this application, it should be understood that the disclosed technical content can be implemented in other ways. The device embodiments described above are merely illustrative; for example, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection of units or modules may be electrical or other forms.

[0120] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0121] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0122] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as a USB flash drive, read-only memory (ROM), random access memory (RAM), portable hard drive, magnetic disk, or optical disk.

[0123] The above description is only a preferred embodiment of this application. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the principle of this application, and these improvements and modifications should also be considered within the scope of protection of this application.

Claims

1. A method for handling financial risk, characterized in that, include: With the authorization of the target user, obtain the target user's financial data within a preset time period, wherein the financial data refers to the business data associated with the target user when conducting financial business within the preset time period; A risk identification model is extracted from the blockchain, the financial data is input into the risk identification model, and the risk identification result is obtained. The risk identification model is obtained by aggregating risk identification sub-models trained by M participating objects. The risk identification model is used to perform time series analysis on the financial data to obtain time series features, evaluate the time series features to obtain a risk score, and generate the risk identification result based on the risk score. M is a positive integer. Based on the risk identification results, a risk handling strategy is determined, and the financial account of the target user is processed according to the risk handling strategy.

2. The method according to claim 1, characterized in that, Obtaining the financial data of the target user within a preset time period includes: Receive a data collection instruction, and collect the target user's initial financial data from the blockchain according to the data collection instruction; The initial financial data is cleaned by obtaining the cleaning rules from the blockchain and then cleaned based on the cleaning rules to obtain the cleaned financial data. A preset encryption algorithm is obtained, an encryption public key is determined based on the preset encryption algorithm, and the cleaned financial data is encrypted using the encryption public key to obtain the financial data.

3. The method according to claim 1, characterized in that, The risk identification model is obtained through the following steps: Financial training samples associated with the M participants within a historical time period are obtained from the blockchain to obtain M sets of financial training samples. Each set of financial training samples is obtained by processing with a preset encryption algorithm. Each set of financial training samples includes N sets of financial training data and the risk level corresponding to each set of financial training data, where N is a positive integer. The preset risk identification model is trained using the M sets of financial training samples to obtain M risk identification sub-models, and the risk identification model is obtained by aggregating the M risk identification sub-models.

4. The method according to claim 3, characterized in that, The risk identification model is obtained by aggregating the M risk identification sub-models, including: Obtain the model parameters of each risk identification sub-model to get M sets of model parameters, encapsulate the M sets of model parameters into a data packet, and broadcast the data packet to Y blockchain nodes deployed in the blockchain, where Y is a positive integer; When the Y blockchain nodes receive the data packet, the Y blockchain nodes perform consensus processing on the data packet to obtain a consensus result; If the Y blockchain nodes accept the data packet without error, the step of obtaining the risk identification model by aggregating the M risk identification sub-models is executed.

5. The method according to claim 4, characterized in that, The consensus results obtained by the Y blockchain nodes through consensus processing of the data packets include: A consensus algorithm is obtained, and each blockchain node sequentially verifies the M sets of model parameters in the data packet according to the consensus algorithm to obtain M verification results; If the verification results of K blockchain nodes indicate that any one of the M sets of model parameters does not match the preset model rules, the Y blockchain nodes will vote on the M sets of model parameters, where K is less than Y / 2 and K is a positive integer. If the data packet is approved by the Y blockchain nodes, it is confirmed that the packet is correct.

6. The method according to claim 3, characterized in that, The M risk identification sub-models are aggregated to obtain the risk identification model in the following manner: The candidate risk identification model is obtained by aggregating the M risk identification sub-models; A preset protocol is invoked, and the candidate risk identification model is detected based on the preset protocol. If the candidate risk identification model fails the detection, the step of training the preset risk identification model using the M sets of financial training samples is executed. If the candidate risk identification model passes the test, the Y blockchain nodes deployed in the blockchain will perform consensus processing on the candidate risk identification model. If the Y blockchain nodes accept the candidate risk identification model, the candidate risk identification model will be determined as the risk identification model.

7. The method according to claim 1, characterized in that, Before inputting the financial data into the risk identification model, the method further includes: The financial data is processed using a homomorphic decryption operation to obtain the initial decrypted data; Receive a key instruction, extract the encryption private key from the key instruction, and decrypt the initial decryption data according to the encryption private key to obtain the decrypted financial data; Based on the decrypted financial data, the step of inputting the financial data into the risk identification model is performed.

8. A financial risk processing device, characterized in that, include: The acquisition unit is used to acquire the financial data of the target user within a preset time period when the target user authorizes it, wherein the financial data refers to the business data associated with the target user when conducting financial business within the preset time period. An extraction unit is used to extract a risk identification model from the blockchain, input the financial data into the risk identification model, process it to obtain a risk identification result, wherein the risk identification model is obtained by aggregating risk identification sub-models trained by M participating objects, the risk identification model is used to perform time series analysis on the financial data to obtain time series features, evaluate the time series features to obtain a risk score, and generate the risk identification result based on the risk score, where M is a positive integer; The determining unit is used to determine a risk handling strategy based on the risk identification result, and to process the financial account of the target user according to the risk handling strategy.

9. An electronic device, characterized in that, include: Memory, which stores executable programs; A processor for running the program, wherein the program, when running, performs the financial risk processing method according to any one of claims 1 to 7.

10. A computer program product comprising computer instructions, characterized in that, When the computer instructions are executed by the processor, they implement the steps of the financial risk handling method according to any one of claims 1 to 7.

Images