A big data analysis system and method for a financial guarantee circle

Abstract

The application discloses a financial guarantee circle big data analysis system and method, relates to the technical field of data analysis, and comprises the following steps: collecting industrial and commercial, judicial and financial data in real time through an API gateway, constructing a quantum guarantee network model and a quantum annealing parameter configuration file after preprocessing, executing implicit relation detection by using a quantum annealing machine, generating a high-correlation guarantee circle list and a quantum calculation log in combination with Bayesian confidence calibration, simulating a risk transmission path based on a synapse plasticity model of a pulse neural network, outputting a synapse weight change table and a dynamic transmission graph, mapping a risk level to a hierarchical blocking strategy matrix through a dynamic blocking decision engine, performing topology reinforcement by using a minimum spanning tree algorithm in combination with historical data, generating a stable subnet list and a protection path graph, and finally weighting and fusing quantum logs, transmission models and subnet data to construct a three-dimensional risk atlas and optimizing real-time decision by using a long short-term memory network and an attention mechanism network.

Description

Technical Field

[0001] The present invention relates to the field of data analysis technology, and in particular to a big data analysis system and method for a financial guarantee circle. Background Art

[0002] With the deepening of global economic integration, the complexity and uncertainty of financial markets are increasing, rendering traditional risk management methods inadequate when dealing with complex guarantee networks. A financial guarantee network is a network structure formed by a series of mutually guaranteeing relationships. Its internal complexity lies not only in the large number of participating entities but also in the multi-layered and multi-dimensional relationships between these entities. Early research focused on analyzing individual or a small number of guarantee relationships, employing statistical and traditional graph theory-based approaches to assess risk. However, with the advent of the big data era, financial institutions have accumulated vast amounts of data resources, including business registration information, judicial decision records, and financial market transaction data. This has opened up the possibility of a deeper understanding and prediction of risks in financial guarantee networks. Against this backdrop, how to effectively utilize these rich data resources to build a system that can reflect market dynamics in real time and accurately predict potential risks has become a hot topic of research.

[0003] While existing technologies have made some progress, they still have many shortcomings in addressing financial guarantee circles. First, traditional data analysis methods struggle to effectively process large and rapidly changing datasets, especially when faced with complex data structures such as quantized guarantee network models and quantum annealing parameter profiles. Second, existing technologies are limited in their ability to detect implicit relationships within guarantee circles, making it impossible to accurately identify highly correlated guarantee circles, thereby affecting the accuracy of risk assessment. Furthermore, current risk transmission modeling mostly relies on static models, lacking effective simulation of dynamic processes and failing to timely reflect the true path of risk transmission. Summary of the Invention

[0004] In view of the above existing problems, the present invention is proposed.

[0005] Therefore, the present invention provides a method for big data analysis of financial guarantee circles to solve the problems faced by existing technologies in processing complex financial guarantee networks, such as low efficiency, inaccurate identification of implicit relationships, and insufficient dynamic simulation of risk transmission paths.

[0006] In order to solve the above technical problems, the present invention provides the following technical solutions:

[0007] In a first aspect, the present invention provides a method for big data analysis of financial guarantee circles, which includes collecting real-time business data of industrial and commercial, judicial and financial categories through an API gateway in real time, and converting the real-time business data into a quantized guarantee network model and a quantum annealing parameter configuration file through preprocessing; according to the quantized guarantee network model and the quantum annealing parameter configuration file, through quantum implicit relationship detection and confidence calibration, a list of potential high-correlation guarantee circles and a quantum computing result log are obtained; according to the list of potential high-correlation guarantee circles and the real-time business data, through synaptic plasticity risk conduction modeling of a pulse neural network, a synaptic weight change record table and a risk conduction path dynamic graph are generated; using a dynamic adaptive blocking decision engine, the synaptic weight change record table and the risk conduction path graph are automatically matched into a hierarchical blocking strategy matrix; external historical risk data is collected in real time, and combined with the hierarchical blocking strategy matrix, a stable core subnet list and a topology protection path distribution map are obtained through topology robustness reinforcement; weighted fusion is performed based on the quantum computing result log, risk conduction modeling, the stable core subnet list and the topology protection path distribution map to generate a three-dimensional risk map, and a deep network optimizer is constructed.

[0008] As a preferred solution of the method for big data analysis of the financial guarantee circle described in the present invention, wherein: the real-time business data is converted into a quantized guarantee network model and a quantum annealing parameter configuration file through preprocessing, the specific steps are as follows:

[0009] Clean and format real-time business data;

[0010] Based on the pre-processed implementation business data, the industry average is obtained according to the industry to which the enterprise belongs, and a preliminary risk score is calculated for each enterprise;

[0011] The calculated preliminary risk score is passed through a quantum-enhanced dynamic confidence verification mechanism to form an enterprise risk score table;

[0012] Through the enterprise risk scoring table, the enterprise ID pairs and coupling coefficients are combined to form a guarantee relationship quantitative matrix;

[0013] Based on the hardware architecture of the quantum annealer, the guarantee relationship quantization matrix is ​​embedded in the physical quantum bit connection to generate a quantized guarantee network model;

[0014] Set the annealing schedule based on the timing of the quantum annealing process and set the flux bias value for each qubit based on the requirements of the optimization problem;

[0015] The guarantee relationship quantization matrix, physical quantum bits, annealing schedule and flux bias value are integrated to form a quantum annealing parameter profile.

[0016] As a preferred solution of the method for big data analysis of the financial guarantee circle described in the present invention, the specific steps of the quantum implicit relationship detection and confidence calibration are as follows:

[0017] Set the annealing parameters on the D-Wave quantum annealer according to the annealing time and temperature in the quantum annealing parameter configuration file;

[0018] Use Bayesian inference to update the prior probability of the highly correlated guarantee circles identified in the annealing parameters and obtain the posterior probability distribution;

[0019] Write a script to automatically collect the key decision points and calculation results for setting annealing parameters and posterior probability distribution, obtain a list of potential highly correlated guarantee circles, and generate a detailed log file.

[0020] As a preferred solution of the method for big data analysis of the financial guarantee circle of the present invention, the synaptic plasticity risk conduction modeling of the pulse neural network is specifically performed as follows:

[0021] Use pulse neural networks to simulate synaptic plasticity, capture the risk propagation path in the guarantee network, record each weight change, and establish a synaptic weight change record table;

[0022] Based on the synaptic weight change record table, a dynamic diagram of the risk conduction path is generated through graphical tools.

[0023] As a preferred solution of the method for big data analysis of financial guarantee circles of the present invention, the dynamic adaptive blocking decision engine has the following specific steps:

[0024] Self-organizing and generating a dynamic and adaptive blocking decision engine, automatically classifying risk levels based on the prominent weight change record table and risk transmission path diagram;

[0025] Formulate a graded blocking strategy based on the risk level and convert it into a matrix form to form a graded blocking strategy matrix.

[0026] As a preferred solution of the method for big data analysis of financial guarantee circles described in the present invention, the topology robustness reinforcement is specifically performed in the following steps:

[0027] Collect external historical risk data and combine it with a hierarchical blocking strategy matrix to identify stable core subnets;

[0028] Use the minimum spanning tree algorithm to design a topology protection path distribution map based on the stable core subnet.

[0029] As a preferred solution of the method for big data analysis of financial guarantee circles described in the present invention, the weighted fusion has the following specific steps:

[0030] The quantum computing result log, risk transmission modeling, stable core subnet list and topology protection path distribution map are weighted and integrated to generate a three-dimensional risk map.

[0031] Use reinforcement learning algorithms to continuously optimize the network structure of the three-dimensional risk map and build a deep network optimizer.

[0032] In a second aspect, the present invention provides a system for big data analysis of financial guarantee circles, comprising:

[0033] The quantum processing module collects real-time business data from industrial and commercial, judicial, and financial sectors through the API gateway, and converts the real-time business data into a quantum guarantee network model and quantum annealing parameter configuration file through pre-processing;

[0034] The quantum circle detection module obtains a list of potential highly correlated guarantee circles and a quantum computing result log based on the quantized guarantee network model and quantum annealing parameter configuration file through quantum implicit relationship detection and confidence calibration;

[0035] The pulse risk control module uses a spiking neural network to model synaptic plasticity risk conduction based on a list of potential highly correlated guarantee circles and real-time business data, generating a synaptic weight change record table and a dynamic diagram of the risk conduction path.

[0036] The dynamic blocking module uses a dynamic adaptive blocking decision engine to automatically match the synaptic weight change record table and risk conduction path diagram into a hierarchical blocking strategy matrix;

[0037] The topology reinforcement module collects external historical risk data in real time and combines it with a hierarchical blocking strategy matrix to obtain a stable core subnet list and a topology protection path distribution map through topology robustness reinforcement.

[0038] The three-dimensional optimization module performs weighted fusion based on quantum computing result logs, risk conduction modeling, stable core subnet lists, and topology protection path distribution maps to generate a three-dimensional risk map and build a deep network optimizer.

[0039] In a third aspect, the present invention provides a computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: when the computer program is executed by the processor, it implements any step of the method for big data analysis of financial guarantee circles as described in the first aspect of the present invention.

[0040] In a fourth aspect, the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein: when the computer program is executed by a processor, it implements any step of the method for big data analysis of financial guarantee circles as described in the first aspect of the present invention.

[0041] The present invention has the following beneficial effects: through the dynamic adaptive blocking decision engine step, it realizes the function of automatically adjusting risk management strategies based on the latest risk information, avoiding the delays and errors caused by human intervention, enhancing the stability and risk resistance of the system, providing an efficient risk management tool, and dynamically adjusting the hierarchical blocking strategy according to actual conditions, further improving the adaptability and robustness of the system. These steps work together to significantly improve the effectiveness and accuracy of big data analysis in the financial guarantee circle. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0043] Figure 1 This is a flowchart of the method for big data analysis of the financial guarantee circle in Example 1.

[0044] Figure 2 This is a flowchart of the layered architecture of the big data analysis of the financial guarantee circle in Example 1.

[0045] Figure 3 This is a data flow perspective flowchart for the big data analysis of the financial guarantee circle in Example 1.

[0046] Figure 4 This is a module perspective flow chart of the big data analysis of the financial guarantee circle in Example 1. DETAILED DESCRIPTION

[0047] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific embodiments of the present invention are described in detail below with reference to the accompanying drawings.

[0048] In the following description, many specific details are set forth to facilitate a full understanding of the present invention. However, the present invention may also be implemented in other ways different from those described herein. Those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0049] Secondly, the term "one embodiment" or "embodiment" herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The phrase "in one embodiment" appearing in various places throughout this specification does not necessarily refer to the same embodiment, nor does it refer to a separate or selective embodiment that is mutually exclusive of other embodiments.

[0050] Example 1, with reference to Figures 1 to 4, which is the first embodiment of the present invention, provides a method for big data analysis of a financial guarantee circle, comprising the following steps:

[0051] S1: Real-time business data of industrial and commercial, judicial and financial categories are collected through the API gateway, and the real-time business data is converted into a quantum guarantee network model and quantum annealing parameter configuration file through preprocessing.

[0052] The business data collected in real time is removed of incomplete and erroneous data and the format is unified; based on the preprocessed data, the industry average is obtained according to the industry to which the enterprise belongs, and a preliminary risk score is calculated for each enterprise; the calculated preliminary risk score is manually reviewed to form an enterprise risk score table; combined with the enterprise risk score table, the enterprise ID pairs and coupling coefficients are formed into a guarantee relationship quantization matrix; based on the hardware architecture of the quantum annealing machine, the guarantee relationship quantization matrix is ​​embedded in the physical quantum bit connection to generate a quantized guarantee network model; according to the time schedule of the quantum annealing process, an annealing schedule is set, and according to the requirements of the optimization problem, a flux bias value is set for each quantum bit; the guarantee relationship quantization matrix, physical quantum bits, annealing schedule and flux bias value are integrated to form a quantum annealing parameter configuration file.

[0053] Select and integrate a reliable API gateway service, such as Alibaba Cloud API Gateway, to subscribe to industrial and commercial, judicial, and financial data sources. Complete the authentication and authorization process according to the data provider's requirements to ensure stable reception of real-time updated business data.

[0054] Deploy an Apache Kafka cluster (a distributed streaming data platform data system) locally or in the cloud. Ensure that the Apache Kafka cluster has sufficient performance to handle high-throughput data streams. Configure producers to send data received from the API gateway to the message classification channel in Apache Kafka, and consumers to read data from the message classification channel for subsequent processing.

[0055] Use Apache Flink to build a real-time data processing pipeline and perform pre-processing operations such as cleaning and transformation on the data, for example, filtering out incomplete or erroneous data records.

[0056] Based on business needs, a specific guarantee network model is self-organized and generated. The guarantee network model should include but not be limited to key fields such as guarantor ID, guaranteed ID, and guarantee amount, and define the guarantee relationship strength, expressed as:

[0057] ;

[0058] in Indicates guarantor With the guarantor The strength of the guarantee relationship between Indicates guarantor To the guarantor The amount of the guarantee, is the sum of all collateral amounts in the entire network. This formula helps quantify the importance of collateral relationships.

[0059] The preprocessed data is mapped to a predefined quantized guarantee network model. At the same time, the parameter configuration file required by the quantum annealing algorithm is generated according to the characteristics of the guarantee network model.

[0060] S2: Based on the quantized guarantee network model and quantum annealing parameter configuration file, through quantum implicit relationship detection and confidence calibration, a list of potential highly correlated guarantee circles and a quantum computing result log are obtained.

[0061] According to the return time and temperature in the quantum annealing parameter configuration file, appropriate annealing parameters are set on the D-Wave quantum annealer; Bayesian inference is used to update the prior probability of the highly correlated guarantee circles identified in the annealing parameters and obtain the posterior probability distribution; a script is written to automatically collect the key decision points and calculation results for setting the annealing parameters and the posterior probability distribution, obtain a list of potential highly correlated guarantee circles, and generate a detailed log file.

[0062] The obtained quantized guarantee network model and the corresponding quantum annealing parameter configuration file are uploaded to the D-Wave quantum computing platform using the API or SDK provided by D-Wave.

[0063] On the D-Wave platform, we select an appropriate quantum algorithm (such as the QUBO problem transformation method) to optimize and solve the correlations within the guarantee network. Based on the characteristics of the quantum annealing process, we set the annealing time and other parameters, execute the quantum computing task, and obtain the potential high-correlation relationships between nodes in the guarantee circle. Based on the quantum computing results, we use Bayesian theorem to adjust the confidence level of the list of potential high-correlation guarantee circles.

[0064] S3: Based on the list of potential highly correlated guarantee circles and real-time business data, the synaptic plasticity risk conduction modeling of the pulse neural network is used to generate a synaptic weight change record table and a dynamic diagram of the risk conduction path.

[0065] The specific steps are as follows: use a pulse neural network to simulate synaptic plasticity, capture the risk propagation path in the guarantee network, record each weight change, and establish a synaptic weight change record table; based on the synaptic weight change record table, generate a dynamic diagram of the risk transmission path through a graphical tool.

[0066] Nodes and their connection relationships are extracted from the obtained list of potential highly correlated guarantee circles, and an SNN (spiking neural network constructed from a list of potential highly correlated guarantee circles) model is designed based on the nodes and connection relationships. Each node represents a guarantee entity, and the guarantee relationship is represented by directed edges.

[0067] The Hebbian learning rule is applied as the change rule of synaptic weights in the SNN model. According to the activity level and learning rate of the previous and next guarantee entities, a function is implemented to receive new risk information in real time and dynamically update the highlight weights.

[0068] During the change process, for each time step, all fields related to the synaptic weight, such as the timestamp, synaptic identifier, and synaptic weight value, are recorded, and the data storage format is modified to the format for subsequent analysis and visualization.

[0069] In the above content, the learning rate is an important parameter in the field of machine learning and neural networks. It is a positive number between 0 and 1, which is used to adjust the size of each update step during the SNN model training process.

[0070] S4: A dynamic adaptive blocking decision engine is used to automatically match the synaptic weight change record table and risk conduction path diagram into a hierarchical blocking strategy matrix.

[0071] The specific steps are as follows: driven by real-time data stream, the pulse neural network is used to dynamically encode the changes in the input synaptic weights, and the risk conduction path diagram is converted into a multi-dimensional tensor of quantum entangled state; the quantum annealing optimizer is used to complete the topological matching of the risk conduction path and the blocking strategy within 20 milliseconds to generate the initial strategy matrix; the strategy forest based on reinforcement learning is used to iteratively optimize the matrix at the millisecond level, and the blocking strength coefficient is dynamically adjusted through a dual-channel self-verification mechanism to complete the self-organization generation of a dynamic adaptive blocking decision engine.

[0072] Design a dynamic and adaptive blocking decision engine to automatically divide risk levels based on the prominent weight change record table and risk transmission path diagram; formulate graded blocking strategies based on risk levels and convert them into matrix form to form a graded blocking strategy matrix.

[0073] Obtain historical data from the synaptic weight change record table and risk conduction path dynamic graph, and calculate the importance measure of each node , the formula is as follows,

[0074] ;

[0075] in, Representation node To all other nodes The shortest path length, is the total number of nodes in the network.

[0076] Using machine learning methods, the hierarchical blocking strategy matrix is ​​passed through a temporal convolutional network to capture timing fluctuations, a graph attention network to model the risk propagation path, and a gated recurrent unit to fuse spatiotemporal features to generate probability, thus obtaining a risk probability model.

[0077] The dynamic data features in real-time data are brought into the risk probability model, and a node high-risk data table is obtained through full-node reasoning to predict the probability of each node becoming a high-risk point in the future.

[0078] Financial risk management is viewed as a Markov decision process. The state space consists of nodes and risk scores, and the action space includes different levels of monitoring and intervention measures taken on specific nodes. Rewards are defined based on the reduced risk value. The deep Q network (DQN) is used as the core algorithm to improve learning efficiency through the experience replay mechanism.

[0079] Through the DQN model, the optimal action sequence is output for each node risk combination to form a hierarchical blocking strategy matrix. This is then tested in a simulation environment, and the model parameter optimization strategy is adjusted based on the test feedback. The final hierarchical blocking strategy matrix is ​​applied to the actual system to guide risk management decisions in real time.

[0080] S5: Collect external historical risk data in real time, and combine it with the hierarchical blocking strategy matrix to obtain a stable core subnet list and topology protection path distribution map through topology robustness reinforcement.

[0081] The specific steps are as follows: collect external historical risk data, combine it with the hierarchical blocking strategy matrix, and identify stable core subnets; use the minimum spanning tree algorithm to design a topological protection path distribution map based on the stable core subnet.

[0082] Based on public financial databases and government-released economic reports, external historical risk data is collected that meets the requirements of authority of the publishing entity, strict quality control, and strong verifiability. ETL tools are used to clean and transform the data so that the format is compatible with existing risk transmission models.

[0083] Based on the integrated risk transmission model data, a financial network diagram is constructed using graph theory software. The node degree centrality of each financial network diagram is calculated using the following formula:

[0084] ;

[0085] in, Representation node The symbol of degree centrality is, Value Node The degree, Indicates the total number of nodes in the network.

[0086] Based on the calculated centrality index, the core nodes in the network are identified, and through the community detection algorithm, the densely connected areas in the network are identified as stable core subnetworks.

[0087] According to the hierarchical blocking strategy matrix, the identified temperature core subnet is optimized and adjusted, and simulation tools are used to simulate the impact of different levels of risk time on the core subnet. The robustness of the core subnet is evaluated, and the network structure parameters are adjusted according to the simulation results to guide the achievement of the optimal robustness level.

[0088] The shortest path algorithm is used to plan multiple protection paths for each key node, and the protection paths are implemented in the actual network, including security enhancement measures at the physical and logical layers. The protection paths are then drawn using graphical tools to form a topological protection path distribution map.

[0089] S6: Based on the quantum computing result log, risk conduction modeling, stable core subnet list and topology protection path distribution map, weighted fusion is performed to generate a three-dimensional risk map and build a deep network optimizer.

[0090] The specific steps are as follows: weighted fusion of quantum computing result logs, risk conduction modeling, stable core subnet lists, and topology protection path distribution maps to generate a three-dimensional risk map; use reinforcement learning algorithms to continuously optimize the network structure and build a deep network optimizer.

[0091] According to different data sources and types, appropriate feature extraction methods are applied, such as PCA principal component analysis and t-SNE dimensionality reduction technology, to convert multi-dimensional data into feature vectors suitable for three-dimensional visualization.

[0092] Use 3D visualization tools such as Unity3D or Three.js to construct a 3D risk map based on feature vectors, and develop interactive analysis and filter functions to allow users to explore specific information in the map by clicking, dragging, etc.

[0093] Build a deep network optimizer based on TensorFlow or PyTorch, define the neural network architecture, including the input layer, multiple hidden layers, and output layer, design a deep learning framework, and build a long short-term memory network (LSTM) model.

[0094] The LSTM model is trained using a historical risk dataset, and supervised learning methods are used to optimize parameters.

[0095] The integrated real-time data stream processing capability enables the deep network optimizer to receive and process the latest business data, automatically trigger early warning signals based on the results output by the LSTM model, and provide decision-making recommendations.

[0096] This embodiment also provides a system for big data analysis of financial guarantee circles, including:

[0097] The quantum processing module collects real-time business data from industrial and commercial, judicial, and financial sectors through the API gateway, and converts the real-time business data into a quantum guarantee network model and quantum annealing parameter configuration file through pre-processing;

[0098] The quantum circle detection module obtains a list of potential highly correlated guarantee circles and a quantum computing result log based on the quantized guarantee network model and quantum annealing parameter configuration file through quantum implicit relationship detection and confidence calibration;

[0099] The pulse risk control module uses a spiking neural network to model synaptic plasticity risk conduction based on a list of potential highly correlated guarantee circles and real-time business data, generating a synaptic weight change record table and a dynamic diagram of the risk conduction path.

[0100] The dynamic blocking module uses a dynamic adaptive blocking decision engine to automatically match the synaptic weight change record table and risk conduction path diagram into a hierarchical blocking strategy matrix;

[0101] The topology reinforcement module collects external historical risk data in real time and combines it with a hierarchical blocking strategy matrix to obtain a stable core subnet list and a topology protection path distribution map through topology robustness reinforcement.

[0102] The three-dimensional optimization module performs weighted fusion based on quantum computing result logs, risk conduction modeling, stable core subnet lists, and topology protection path distribution maps to generate a three-dimensional risk map and build a deep network optimizer.

[0103] This embodiment also provides a computer device, which is suitable for the method of big data analysis of financial guarantee circles, including: a memory and a processor; the memory is used to store computer-executable instructions, and the processor is used to execute computer-executable instructions to implement the method of big data analysis of financial guarantee circles proposed in the above embodiment.

[0104] The computer device can be a terminal, and the computer device includes a processor, a memory, a communication interface, a display screen and an input device connected by a system bus. The processor of the computer device is configured to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for running the operating system and the computer program in the non-volatile storage medium. The communication interface of the computer device is configured to perform wired or wireless communication with an external terminal. The wireless communication can be achieved by WIFI, an operator network, NFC (Near Field Communication) or other technologies. The display screen of the computer device can be a liquid crystal display screen or an electronic ink display screen. The input device of the computer device can be a touch layer overlaid on the display screen, or a key, trackball or touchpad arranged on the shell of the computer device, or an external keyboard, touchpad or mouse, etc.

[0105] The embodiment also provides a storage medium having a computer program stored thereon, and the computer program is executed by a processor to implement the method for big data analysis of a financial guarantee circle as described in the above embodiment. The storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as a static random access memory (SRAM), an electrically erasable programmable read-only memory (EEPROM), an erasable programmable read-only memory (EPROM), a programmable read-only memory (PROM), a read-only memory (ROM), a magnetic memory, a flash memory, a magnetic disk or an optical disk.

[0106] To sum up, the application achieves the function of automatically adjusting the risk management strategy based on the latest risk information through the dynamic adaptive blocking decision engine step, avoids the delay and error caused by human intervention, enhances the stability and anti-risk ability of the system, provides an efficient risk management tool, and dynamically adjusts the hierarchical blocking strategy according to the actual situation, further improves the adaptability and robustness of the system. These steps work together to significantly improve the effectiveness and accuracy of the big data analysis of the financial guarantee circle.

[0107] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A method for big data analysis of a financial guarantee circle, characterized by: include, Real-time business data from industry and commerce, judicial affairs, and finance are collected through the API gateway, and converted into quantum guarantee network models and quantum annealing parameter configuration files through preprocessing. According to the quantized guarantee network model and quantum annealing parameter configuration file, through quantum implicit relationship detection and confidence calibration, a list of potential highly correlated guarantee circles and a quantum computing result log are obtained; Based on the list of potential highly correlated guarantee circles and real-time business data, the synaptic plasticity risk conduction modeling of the spiking neural network is used to generate a synaptic weight change record table and a dynamic diagram of the risk conduction path; A dynamic adaptive blocking decision engine is used to automatically match the synaptic weight change record table and risk conduction path diagram into a hierarchical blocking strategy matrix; Real-time collection of external historical risk data, combined with a hierarchical blocking strategy matrix, and topology robustness reinforcement to obtain a stable core subnet list and topology protection path distribution map; Based on the quantum computing result log, risk conduction modeling, stable core subnet list and topology protection path distribution map, weighted fusion is performed to generate a three-dimensional risk map and build a deep network optimizer.

2. The method for big data analysis of financial guarantee circles according to claim 1, characterized in that: The preprocessing process converts real-time business data into a quantized guarantee network model and a quantum annealing parameter configuration file. The specific steps are as follows: Clean and format real-time business data; Based on the pre-processed implementation business data, the industry average is obtained according to the industry to which the enterprise belongs, and a preliminary risk score is calculated for each enterprise; The calculated preliminary risk score is passed through a quantum-enhanced dynamic confidence verification mechanism to form an enterprise risk score table; Through the enterprise risk scoring table, the enterprise ID pairs and coupling coefficients are combined to form a guarantee relationship quantitative matrix; Based on the hardware architecture of the quantum annealer, the guarantee relationship quantization matrix is ​​embedded in the physical quantum bit connection to generate a quantized guarantee network model; Set the annealing schedule based on the timing of the quantum annealing process and set the flux bias value for each qubit based on the requirements of the optimization problem; The guarantee relationship quantization matrix, physical quantum bits, annealing schedule and flux bias value are integrated to form a quantum annealing parameter profile.

3. The method for big data analysis of financial guarantee circles according to claim 1, characterized in that: The specific steps of quantum hidden relationship detection and confidence calibration are as follows: Set the annealing parameters on the D-Wave quantum annealer according to the annealing time and temperature in the quantum annealing parameter configuration file; Use Bayesian inference to update the prior probability of the highly correlated guarantee circles identified in the annealing parameters and obtain the posterior probability distribution; Write a script to automatically collect the key decision points and calculation results for setting annealing parameters and posterior probability distribution, obtain a list of potential highly correlated guarantee circles, and generate a detailed log file.

4. The method for big data analysis of financial guarantee circles according to claim 1, characterized in that: The synaptic plasticity risk conduction modeling of the pulse neural network is carried out in the following specific steps: Use pulse neural networks to simulate synaptic plasticity, capture the risk propagation path in the guarantee network, record each weight change, and establish a synaptic weight change record table; Based on the synaptic weight change record table, a dynamic diagram of the risk conduction path is generated through graphical tools.

5. The method for big data analysis of financial guarantee circles according to claim 1, characterized in that: The dynamic adaptive blocking decision engine has the following specific steps: Self-organizing and generating a dynamic and adaptive blocking decision engine, automatically classifying risk levels based on the prominent weight change record table and risk transmission path diagram; Formulate a graded blocking strategy based on the risk level and convert it into a matrix form to form a graded blocking strategy matrix.

6. The method for big data analysis of financial guarantee circles according to claim 1, characterized in that: The topology robustness reinforcement is carried out in the following specific steps: Collect external historical risk data and combine it with a hierarchical blocking strategy matrix to identify stable core subnets; Use the minimum spanning tree algorithm to design a topology protection path distribution map based on the stable core subnet.

7. The method for big data analysis of financial guarantee circles according to claim 1, characterized in that: The specific steps of weighted fusion are as follows: The quantum computing result log, risk transmission modeling, stable core subnet list and topology protection path distribution map are weighted and integrated to generate a three-dimensional risk map. Use reinforcement learning algorithms to continuously optimize the network structure of the three-dimensional risk map and build a deep network optimizer.

8. A system for big data analysis of financial guarantee circles, based on the method for big data analysis of financial guarantee circles according to any one of claims 1 to 7, characterized in that: include, The quantum processing module collects real-time business data from industrial and commercial, judicial, and financial sectors through the API gateway, and converts the real-time business data into a quantum guarantee network model and quantum annealing parameter configuration file through pre-processing; The quantum circle detection module obtains a list of potential highly correlated guarantee circles and a quantum computing result log based on the quantized guarantee network model and quantum annealing parameter configuration file through quantum implicit relationship detection and confidence calibration; The pulse risk control module uses a spiking neural network to model synaptic plasticity risk conduction based on a list of potential highly correlated guarantee circles and real-time business data, generating a synaptic weight change record table and a dynamic diagram of the risk conduction path. The dynamic blocking module uses a dynamic adaptive blocking decision engine to automatically match the synaptic weight change record table and risk conduction path diagram into a hierarchical blocking strategy matrix; The topology reinforcement module collects external historical risk data in real time and combines it with a hierarchical blocking strategy matrix to obtain a stable core subnet list and a topology protection path distribution map through topology robustness reinforcement. The three-dimensional optimization module performs weighted fusion based on quantum computing result logs, risk conduction modeling, stable core subnet lists, and topology protection path distribution maps to generate a three-dimensional risk map and build a deep network optimizer.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method for big data analysis of the financial guarantee circle described in any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method for big data analysis of the financial guarantee circle described in any one of claims 1 to 7 are implemented.

Images