Cloud secondary chain agent engine system of financial case intelligent risk control platform

Abstract

A cloud secondary chain agent engine system of a financial case intelligent risk control platform constructs a cloud intelligent system architecture integrating a knowledge graph, a neural network model and a block chain technology, performs feature extraction on case information by using the knowledge graph, analyzes the extracted case features by using the neural network model, and performs risk management on financial cases. According to the method, risk level assessment is realized, case judgment is predicted according to a risk assessment result, an electronic document is intelligently generated based on a prediction result and a set rule, and electronic evidence storage is performed on the electronic document and a case processing process by using a block chain technology to ensure that data cannot be tampered, so that the efficiency and the accuracy of financial case processing are improved, and the financial efficiency is improved. And meanwhile, multi-terminal user interaction is supported, so that a user can conveniently input case basic information, track the processing progress in real time and receive suggestions of an intelligent system.

Description

Technical Field

[0001] This invention relates to the intersection of financial technology and artificial intelligence, specifically to a cloud-based intelligent agent engine system for an intelligent risk control platform for financial cases. Background Technology

[0002] This invention provides an intelligent risk control method for financial cases based on the Yunzhonglian intelligent agent architecture engine. It achieves risk identification, adjudication prediction, and process automation for financial cases through multi-agent collaborative technology. The system integrates knowledge graphs, neural network models, and blockchain technology to provide financial institutions with a full-process intelligent risk control solution, including functions such as case feature extraction, risk level assessment, adjudication prediction, automatic document generation, and electronic evidence storage, significantly improving the efficiency and accuracy of financial case processing and risk control. Summary of the Invention

[0003] To address the problems and deficiencies in the prior art, this invention provides a cloud-based intelligent engine system for an intelligent risk control platform for financial cases.

[0004] A cloud-based intelligent risk control method for financial cases, comprising: Construct a cloud-based intelligent system architecture that integrates knowledge graphs, neural network models, and blockchain technologies; Feature extraction of case information is performed using knowledge graphs; A neural network model is used to analyze the extracted case features to achieve risk level assessment; Based on the risk assessment results, the intelligent system architecture is used to predict case decisions. Based on prediction results and established rules, electronic documents are intelligently generated. 1. Utilizing blockchain technology to implement electronic evidence storage for electronic documents and case processing processes to ensure data immutability. 2. In the intelligent risk control method for financial cases as described in claim 1, feature extraction refers to extracting structured case feature values ​​from case-related text, images, transaction records, and other materials. 3. In the intelligent risk control method for financial cases as described in claim 1, risk level assessment is performed using a trained deep learning model. 4. In the intelligent risk control method for financial cases as described in claim 1, case adjudication prediction is a process of dynamically optimizing prediction accuracy based on a dataset of historical cases using reinforcement learning algorithms. 5. In the intelligent risk control method for financial cases as described in claim 1, automatic document generation involves, but is not limited to, template design and filling rules for legal documents such as pleadings, answers, and adjudication recommendations. 6. In the intelligent risk control method for financial cases as described in claim 1, blockchain technology supports a consortium blockchain model and ensures the integrity and traceability of data uploaded to the blockchain. 7. In the intelligent risk control method for financial cases as described in claim 1, it also includes a user interface that allows users to input basic case information, track processing progress in real time, and receive suggestions from the intelligent system. 8. The intelligent risk control method for financial cases as described in claim 7, wherein the user interface supports multiple terminals, is suitable for online and offline operation, and is compatible with desktop and mobile devices.

[0005] Compared with the prior art, the advantages of this invention are as follows: Technology integration By integrating knowledge graph, federated learning and blockchain technologies, it breaks through the limitations of single algorithms.

[0006] Business adaptability It supports financial institutions in risk control throughout the entire lifecycle, from pre-loan review to non-performing asset disposal, improving disposal efficiency by more than 300%.

[0007] Efficiency Improvement Batch processing capabilities reduce the analysis time for tens of thousands of cases from weeks to hours. Detailed Implementation Plan To address the problems and shortcomings of existing technologies in financial case processing, such as low efficiency and insufficient accuracy in risk control, this invention provides an intelligent risk control method for financial cases based on a cloud-based intelligent architecture. This method constructs a cloud-based intelligent architecture integrating knowledge graphs, neural network models, and blockchain technology. It utilizes knowledge graphs to extract features from case information and employs neural network models to analyze the extracted case features, thereby achieving risk level assessment. Based on the risk assessment results, the intelligent architecture predicts case decisions and intelligently generates electronic documents based on the prediction results and established rules. Simultaneously, blockchain technology is used to electronically store the electronic documents and the case processing process to ensure data immutability. Furthermore, the feature extraction refers to extracting structured case feature values ​​from case-related text, images, transaction records, and other materials for risk level assessment. The method utilizes a trained deep learning model, while case judgment prediction is based on a dataset of historical cases, dynamically optimizing prediction accuracy using reinforcement learning algorithms. Automatic document generation includes, but is not limited to, template design and filling rules for legal documents such as pleadings, answers, and judgment recommendations, improving document preparation efficiency and accuracy. The blockchain technology supports a consortium blockchain model, further ensuring the integrity and traceability of on-chain data. The method also includes a user interface that allows users to input basic case information, track processing progress in real time, and receive intelligent system suggestions. This interface supports multiple terminals, is suitable for online and offline operation, and is compatible with desktop and mobile devices, thereby effectively improving the overall efficiency and risk control accuracy of financial case processing.

[0008] The technical solutions in the embodiments of the present invention will be clearly and completely described below. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments in this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application. It should be noted that the terms "first," "second," etc., in the specification and claims of this application are used to distinguish similar objects and are not necessarily used to describe their order. In addition, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or server that includes a series of steps or sub-modules is not necessarily limited to those steps or sub-modules that are explicitly listed, but may include other steps or sub-modules that are not explicitly listed or that are inherent to these processes, methods, products, or devices.

[0009] Example 1: Constructing a cloud-based intelligent system architecture integrating knowledge graphs, neural network models, and blockchain technologies. First, case-related information, including but not limited to text, images, and transaction records, is obtained in real-time from the financial institution's business system. Knowledge graph technology is used to extract features from this data, converting it into structured case feature values. Then, a trained deep learning model is used to assess the risk level of the extracted feature values, with the assessment results categorized into high, medium, and low levels. Based on the risk assessment results, historical case datasets and reinforcement learning algorithms are used to dynamically optimize the prediction of case rulings, generating prediction results. Finally, based on the prediction results and preset legal document templates, electronic documents, such as complaints, answers, and ruling recommendations, are intelligently generated. These electronic documents and their processing are then electronically stored using blockchain technology to ensure data immutability. The entire process is conducted through a graphical user interface, where users can enter basic case information, track processing progress in real-time, and receive suggestions from the intelligent system.

[0010] Example 2: Building upon Example 1, the feature extraction stage is not limited to textual information but also includes the processing of multimodal data such as images and transaction records. A deep learning model is used to comprehensively analyze this multimodal data, improving the comprehensiveness and accuracy of feature extraction. Experimental data shows that compared to risk assessment relying solely on textual information, incorporating images and transaction records improves the accuracy of risk assessment by 15%.

[0011] Example 3: Building upon Example 1, the risk assessment phase employed specific deep learning models, such as Convolutional Neural Networks (CNNs) for image feature extraction and Recurrent Neural Networks (RNNs) for sequence data processing. Through continuous training and optimization of the models, the accuracy of risk assessment reached over 95%. In the test set, the error rate decreased by 20%, demonstrating the effectiveness of the deep learning models.

[0012] Example 4: Building upon Example 1, the case adjudication prediction stage utilizes a large historical case dataset. Reinforcement learning algorithms are used to dynamically adjust the parameters of the prediction model to adapt to the ever-changing market environment. Experimental results show that as the dataset grows, the prediction accuracy steadily improves. In the latest test, the prediction accuracy reached 92%, a 10 percentage point improvement compared to traditional methods, effectively reducing the false positive rate.

[0013] Example 5: Building upon Example 1, the electronic document generation module supports various types of legal document template design and filling rules, including but not limited to pleadings, answers, and recommendations for judgment. Through intelligent template filling technology, the automatically generated documents achieve an accuracy rate of 98%, significantly reducing the workload of manual review. Furthermore, the system supports user-defined templates to meet the needs of different case types.

[0014] Example 6: Building upon Example 1, blockchain technology supports a consortium blockchain model, ensuring the integrity and traceability of all on-chain data. By introducing multiple nodes, the system's transparency and security are enhanced. Experimental data shows that adopting a consortium blockchain reduces the data tampering event rate by 99%, ensuring data security and reliability.

[0015] Example 7: Building upon Example 1, a user interface was added. Users can use this interface to enter basic case information, view real-time case processing progress, and receive suggestions from the intelligent system. The interface supports multiple terminal devices, including desktop computers and mobile devices, meeting diverse user needs. Experimental data shows a 30% increase in user satisfaction, with a particularly significant improvement in the experience on mobile devices, resulting in positive user feedback.

[0016] Example 8: Building upon Example 7, the user interface supports not only online but also offline operation. Users can continue working without a network connection and synchronize data once the network is restored. This feature is particularly suitable for environments with poor network conditions, such as remote areas or during international travel. Experimental data shows that the introduction of the offline operation function enables the system to maintain efficient operation in various environments, further enhancing the user experience.

[0017] It should be noted that the order of the embodiments described above is merely for descriptive purposes and does not represent the superiority or inferiority of the embodiments. In some cases, the actions or steps recorded in the specification and claims can be performed in a different order than that shown in the embodiments, and the desired results can still be achieved. Furthermore, the processes depicted in the drawings do not necessarily require a specific or sequential order to achieve the desired results; in some embodiments, multitasking and parallel processing are also feasible or advantageous. Those skilled in the art will understand that all or part of the steps of the above embodiments can be implemented by hardware, or by a program instructing related hardware to implement them. The program can be stored in a computer-readable storage medium, such as a read-only memory, a disk, or an optical disk. The above are merely preferred embodiments of this application and should not be construed as limiting this application. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the protection scope of this application. Attached Figure Description Figure 1 : Schematic diagram of the financial side technology architecture of Yunzhonglian intelligent agent.

Claims

1. A cloud-based intelligent system architecture-based intelligent risk control method for financial cases, characterized in that, include: Construct a cloud-based intelligent system architecture that integrates knowledge graphs, neural network models, and blockchain technologies; The knowledge graph is used to extract features from case information; The extracted case features are analyzed using the neural network model to achieve risk level assessment; Based on the risk assessment results, the intelligent system architecture is used to predict case decisions. Based on prediction results and established rules, electronic documents are intelligently generated. The aforementioned blockchain technology is used to electronically preserve electronic documents and case processing procedures to ensure the immutability of data.

2. The intelligent risk control method for financial cases as described in claim 1, characterized in that, Feature extraction refers to extracting structured case feature values ​​from case-related materials such as text, images, and transaction records.

3. The intelligent risk control method for financial cases as described in claim 1, characterized in that, The risk level assessment was conducted using a trained deep learning model.

4. The intelligent risk control method for financial cases as described in claim 1, characterized in that, The case judgment prediction is based on a dataset of historical cases, and the process of dynamically optimizing the prediction accuracy using reinforcement learning algorithms.

5. The intelligent risk control method for financial cases as described in claim 1, characterized in that, The automatic document generation involves, but is not limited to, the template design and filling rules for legal documents such as complaints, answers, and ruling recommendations.

6. The intelligent risk control method for financial cases as described in claim 1, characterized in that, The blockchain technology supports the consortium blockchain model and ensures the integrity and traceability of data uploaded to the blockchain.

7. The intelligent risk control method for financial cases as described in claim 1, characterized in that, It also includes a user interface that allows users to enter basic case information, track the processing progress in real time, and receive suggestions from the intelligent system.

8. The intelligent risk control method for financial cases as described in claim 7, characterized in that, The user interface supports multiple terminals, is suitable for online and offline operation, and is compatible with desktop and mobile devices.

Images