Method and system for determining financial risk information, and computer program product

Abstract

The invention discloses a method and a system for determining financial risk information, and a computer program product. Relates to the field of financial science and technology, and the method comprises the steps: determining M financial risk data of a financial institution in a preset time period, and obtaining a case database, the case database comprising a plurality of financial risk cases, and M being a positive integer; matching each piece of financial risk data with a financial risk case in a case database to obtain M groups of rank-related data, inputting the M groups of rank-related data and the M pieces of financial risk data into a risk scoring model, and outputting a risk score; and inputting the risk score into a report generation model, and outputting a financial risk attribution report of the financial institution. Through the risk detection method and device, the technical problem that the detection accuracy is low when risk detection is carried out on a financial institution in the prior art is solved.

Description

Technical Field

[0001] The present application relates to the field of financial technology, and specifically, to a method, system, and computer program product for determining financial risk information. Background Art

[0002] In the financial sector, institutional risk analysis has been widely used as an auxiliary method for risk assessment. Currently, commonly used institutional risk scoring systems are based on the following three technical solutions: First, expert rule-based risk scoring systems, which are based on a pre-defined keyword library and rule library, identify and classify institutional risks through pattern matching, and then quantify the identified risks using a numerical scorecard. However, this approach is limited by a fixed set of rules and vocabulary, and cannot fully understand and process the complex context of financial terminology, industry policy interpretation, and multilingual risks, which can easily lead to biased scoring results.

[0003] Secondly, risk scoring systems based on traditional machine learning employ classic machine learning algorithms such as support vector machines and naive Bayes to perform sentiment classification and risk scoring on recent institutional risk text data. These systems, by learning from large amounts of labeled data, can initially identify positive and negative sentiment within text. However, these approaches are also limited by the model's generalization capabilities and the depth of its semantic understanding. This leads to lower accuracy, particularly when processing specialized information in the financial sector. This impacts the comprehensiveness and timeliness of the scoring results. Furthermore, when evaluating institutions of varying sizes, industries, and sectors, it is difficult to provide a standardized scoring system that is both universally applicable and accounts for individual differences. This makes it difficult for the system to effectively integrate industry context and policy environment into risk analysis, reducing the industry relevance and responsiveness of the scoring.

[0004] There is currently no effective solution to the technical problem of low detection accuracy when conducting risk detection on financial institutions in relevant technologies. Summary of the Invention

[0005] The main purpose of this application is to provide a method, system and computer program product for determining financial risk information to solve the technical problem of low detection accuracy when performing risk detection on financial institutions in related technologies.

[0006] To achieve the above objectives, according to one aspect of the present application, a method for determining financial risk information is provided. The method comprises: determining M financial risk data for a financial institution during a preset time period, and obtaining a case database, wherein the case database includes multiple financial risk cases, and M is a positive integer; matching each piece of financial risk data with the financial risk cases in the case database to obtain M sets of rank-correlated data; inputting the M sets of rank-correlated data and the M financial risk data into a risk scoring model to output a risk score; and inputting the risk score into a report generation model to output a financial risk attribution report for the financial institution.

[0007] Furthermore, determining the M financial risk data of a financial institution in a preset time period includes: obtaining comprehensive risk information for the preset time period and obtaining the institution number of the financial institution; extracting financial risk information associated with the financial institution from the comprehensive risk information according to the institution number to obtain financial risk information to be processed; formatting the financial risk information to be processed to obtain converted financial risk information, and text segmenting the converted financial risk information to obtain M financial risk data.

[0008] Furthermore, each financial risk data is matched with the financial risk cases in the case database to obtain M groups of rank-related data, including: extracting N financial risk cases from the case database, and vectorizing each financial risk case to obtain N candidate case data, where N is a positive integer; for one financial risk data, calculating the semantic similarity between the financial risk data and the N candidate case data respectively, to obtain N similarity data; sorting the N similarity data in descending order to obtain a similarity queue, extracting Y similarity data at a preset position from the similarity queue, and obtaining the candidate case data corresponding to the Y similarity data to obtain Y financial risk case data, where Y is less than N and Y is a positive integer; using the financial risk data and each financial risk case data for splicing to obtain Y rank-related sub-data, and combining the Y rank-related sub-data to obtain a group of rank-related data.

[0009] Furthermore, M groups of rank-related data and M financial risk data are input into a risk scoring model, and a risk score is output, including: adjusting the parameters of the risk scoring model according to risk identification rules to obtain an adjusted risk scoring model, wherein the risk identification rules are determined by the institutional type of the financial institution; for one financial risk data, the adjusted risk scoring model identifies the financial risk factors associated with the financial institution, a group of rank-related data and the financial risk data to obtain K dimension scores, wherein K is a positive integer; determining the dimension weights according to the risk identification rules, and performing weighted calculation on the K dimension scores based on the dimension weights to obtain a risk score.

[0010] Furthermore, financial risk factors are determined in the following manner: obtaining the institutional type of the financial institution, determining risk identification rules based on the institutional type, wherein the risk identification rules include at least: sentiment polarity rules, operational impact rules, and communication rules; for a financial risk data, analyzing a set of rank-related data associated with the financial risk data based on the risk identification rules to obtain financial risk factors, wherein the financial risk factors include at least risk event information and sentiment polarity associated with the risk event information.

[0011] Furthermore, inputting the risk score into the report generation model and outputting the financial risk attribution report of the financial institution includes: obtaining the institutional type of the financial institution and determining a preset report template based on the institutional type; inputting the preset report template and the risk score into the report generation model and outputting the financial risk attribution report.

[0012] Furthermore, the preset report template and risk score are input into the report generation model, and the output of the financial risk attribution report includes: extracting K dimension scores from the risk score, wherein each dimension score is associated with a dimension type, and the dimension types include at least: sentiment polarity dimension, operational impact dimension, and communication scope dimension; obtaining dimension scoring rules, analyzing the K dimension scores based on the dimension scoring rules, and obtaining score explanation information associated with the risk score; inputting the score explanation information, the preset report template, and the risk score into the report generation model, and outputting the financial risk attribution report.

[0013] To achieve the above-mentioned objectives, according to another aspect of the present application, a system for determining financial risk information is provided. The system includes: a database module for obtaining M financial risk data and a case database of a financial institution in a preset time period, and sending the M financial risk data and financial risk cases in the case database to a language processing matching module and a scoring module; a language processing matching module for matching each financial risk data with the financial risk cases in the case database to obtain M groups of rank-related data, and sending the M groups of rank-related data to a scoring module, where M is a positive integer; a scoring module for inputting the M groups of rank-related data and the M financial risk data into a risk scoring model, outputting a risk score, and sending the risk score to a report generation module; and a report generation module for obtaining a preset report template and score explanation information associated with the risk score, inputting the score explanation information, the preset report template, and the risk score into the report generation model, and outputting a financial risk attribution report.

[0014] According to another aspect of an embodiment of the present invention, a computer-readable storage medium is also provided, which includes a stored executable program, wherein when the executable program is running, the device where the computer-readable storage medium is located is controlled to execute any of the above-mentioned methods for determining financial risk information.

[0015] According to another aspect of an embodiment of the present invention, an electronic device is also provided, including one or more processors and a memory, wherein the memory stores an executable program, and the processor is used to run the program, wherein when the one or more programs are executed by one or more processors, the one or more processors implement any of the above-mentioned methods for determining financial risk information.

[0016] According to another aspect of an embodiment of the present invention, a computer program product is further provided. The computer program product includes a computer program, wherein when the computer program is executed by a processor, any one of the above-mentioned methods for determining financial risk information is implemented.

[0017] In an embodiment of the present application, a method of determining financial risk information is adopted, by determining M financial risk data of a financial institution in a preset time period and obtaining a case database, wherein the case database includes multiple financial risk cases, and M is a positive integer; each financial risk data is matched with the financial risk case in the case database to obtain M groups of rank-related data, and the M groups of rank-related data and the M financial risk data are input into a risk scoring model to output a risk score; the risk score is input into a report generation model to output a financial risk attribution report of the financial institution, thereby solving the technical problem of low detection accuracy when performing risk detection on financial institutions in the related technology, by determining the financial risk data and the case database of a financial institution in a preset time period, matching each financial risk data with the financial risk case in the case database to obtain rank-related data, using the risk scoring model to analyze the rank-related data, outputting the risk score, and then using the report generation model to analyze the risk score, and outputting the financial risk attribution report, thereby achieving the technical effect of significantly improving the accuracy and reliability of risk detection on financial institutions. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] The accompanying drawings, which constitute part of this application, are intended to provide a further understanding of this application. The exemplary embodiments and descriptions of this application are intended to explain this application and do not constitute an improper limitation on this application. In the accompanying drawings:

[0019] Figure 1 A hardware structure block diagram of a computer terminal (or mobile device) for implementing a method for determining financial risk information;

[0020] Figure 2 is a flowchart of a method for determining financial risk information provided in an embodiment of the present application;

[0021] Figure 3 This is an example of an optional method for determining financial risk information provided by an embodiment of the present application. Figure 1 ;

[0022] Figure 4is a schematic diagram of a system for determining financial risk information provided according to an embodiment of the present application;

[0023] Figure 5 This is an example of an optional method for determining financial risk information provided by an embodiment of the present application. Figure 2 ;

[0024] Figure 6 is a schematic diagram of a device for determining financial risk information provided in an embodiment of the present application;

[0025] Figure 7 This is a structural block diagram of an electronic device according to an embodiment of the present application. DETAILED DESCRIPTION

[0026] In order to enable those skilled in the art to better understand the present invention, the following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments in the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts should fall within the scope of protection of this application.

[0027] It should be noted that the terms "first", "second", etc. in the specification and claims of the present application and the above-mentioned drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or sequential order. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments of the present application described herein can be implemented in a sequence other than those illustrated or described herein. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device comprising a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0028] It should be noted that the relevant 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 for analysis, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties. For example, an interface is set up between this system and the relevant user or organization. Before obtaining relevant information, it is necessary to send an acquisition request to the aforementioned user or organization through the interface, and obtain the relevant information after receiving the consent information fed back by the aforementioned user or organization.

[0029] It should be noted that the collected information used 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 comply with the relevant laws, regulations and standards of the relevant regions, take necessary confidentiality measures, do not violate public order and good customs, and provide corresponding operation entrances for users to choose to authorize or refuse use.

[0030] Example 1

[0031] According to an embodiment of the present application, an embodiment of a method for determining financial risk information is also provided. It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system 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 can be executed in an order different from that shown here.

[0032] The method embodiment provided in the first embodiment of the present application can be executed in a mobile terminal, a computer terminal or a similar computing device. Figure 1 It is a hardware structure diagram of a computer terminal (or mobile device) for implementing a method for determining financial risk information, such as Figure 1 As shown, the computer terminal 10 (or mobile device) may include one or more ( Figure 1 The computer system 102 includes a processor 102 (the processor 102 may include but is not limited to a microcontroller unit (MCU) or a programmable logic device (FPGA)), a memory 104 for storing data, and a transmission device 106 for communication functions. In addition, the computer system 102 may include: a display, an input / output interface (I / O interface), a universal serial bus (USB) port (which may be included as one of the ports of the BUS), a network interface, a keyboard, a cursor control device, a power supply, and / or a camera. It will be understood by those skilled in the art that Figure 1 The structure shown is only for illustration and does not limit the structure of the above electronic device. Figure 1 More or fewer components than shown, or with Figure 1 Different configurations shown.

[0033] It should be noted that the one or more processors 102 and / or other data processing circuits described above may generally be referred to herein as "data processing circuitry". The data processing circuitry may be embodied in whole or in part as software, hardware, firmware, or any other combination thereof. In addition, the data processing circuitry may be a single independent processing module, or may be incorporated in whole or in part into any of the other components of the computer terminal 10 (or mobile device). As described in the embodiments of the present application, the data processing circuitry serves as a processor control (e.g., selection of a variable resistor terminal path connected to an interface).

[0034] Memory 104 can be used to store software programs and modules for application software, such as the program instructions / data storage device corresponding to the method for determining financial risk information in the embodiments of the present application. Processor 102 executes the software programs and modules stored in memory 104 to perform various functional applications and data processing, thereby implementing the above-mentioned method for determining financial risk information. 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 examples, memory 104 may further include memory remotely located relative to processor 102, and such remote memory may be connected to 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.

[0035] The transmission device 106 is used to receive or send data via a network. A specific example of the aforementioned network may include a wireless network provided by the communications provider of the computer terminal 10. In one embodiment, 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 enable communication with the Internet. In one embodiment, the transmission device 106 may be a radio frequency (RF) module, which is used to communicate with the Internet wirelessly.

[0036] The display may be, for example, a touch screen liquid crystal display (LCD), which enables a user to interact with a user interface of the computer terminal 10 (or mobile device).

[0037] Under the above operating environment, this application provides Figure 2 The method for determining the financial risk information shown. Figure 2 is a flow chart of a method for determining financial risk information according to an embodiment of the present application. Figure 2 As shown, the method includes the following steps:

[0038] Step S201 : determining M financial risk data of a financial institution in a preset time period, and obtaining a case database, wherein the case database includes a plurality of financial risk cases, and M is a positive integer.

[0039] It should be noted that financial risk data can refer to public opinion data from financial institutions. In the risk management of financial institutions, the collection and analysis of public opinion data plays a vital role in predicting market dynamics and assessing investment risks. Public opinion data can reflect the immediate reactions of the market or the public. Therefore, in order to conduct risk assessments on financial institutions, it is first possible to obtain multiple financial risk data from various information sources such as news, social media, market reports, and policy announcements. At the same time, a case database containing basic information about the institution, industry documents, and historical public opinion data can be obtained to facilitate subsequent retrieval and analysis. Each financial risk case in this database includes not only the original text describing the event, but also the cause, process, and outcome of the event, as well as the specific impact on the financial market, financial institutions, and investors, to facilitate comparison and correlation analysis with current financial risk data.

[0040] Step S202 : Match each financial risk data with the financial risk cases in the case database to obtain M groups of rank-related data, and input the M groups of rank-related data and the M financial risk data into a risk scoring model to output a risk score.

[0041] In order to accurately assess financial risks, since financial risks are not only reflected in news reports or public discussions on social media, but also in multi-dimensional data such as industry research, policy interpretation, and historical case analysis, it is possible to construct a risk scoring model that can effectively integrate internal and external information and accurately score financial risks. The risk scoring model can be trained by a large language model (LLM).

[0042] Furthermore, after obtaining the financial risk data and case database of the financial institution, each financial risk data can be matched with the financial risk cases in the case database through the RAG (Retrieval-Augmented Generation) algorithm to obtain multiple sets of rank-related data, that is, chunk vectors. These rank-related data and the corresponding financial risk data are then input into the above-mentioned risk scoring model. After receiving the input, the LLM model first extracts features (such as polarity, intensity, duration, and spread range), then analyzes the case data that matches the data, and finally combines the results of the feature and case analysis to generate a comprehensive risk score that can reflect the overall risk faced by the current financial institution in the current market environment. The generation of the score takes into account the influence of all features and their weighted effects, as well as the comprehensive judgment of the information in the case database.

[0043] Step S203: input the risk score into the report generation model and output the financial risk attribution report of the financial institution.

[0044] Specifically, after generating a risk score that reflects a financial institution's quantitative assessment of public opinion, the market, and potential risk factors, this score can be input into a report generation model, transforming it into a specific, actionable financial risk attribution report. For example, the model can deeply analyze the risk score to determine which specific events or factors triggered the volatility and how they affected the financial institution's risk score. Based on the attribution analysis and contextual integration of the risk score, targeted risk control recommendations can be provided. Finally, the model outputs a financial risk attribution report that includes a risk overview, key risk factors, external environment analysis, risk scenario simulations, and operational recommendations and strategic planning. This report, in turn, provides strong support for the development of timely and effective risk management strategies.

[0045] The method for determining financial risk information provided in the embodiment of the present application determines M financial risk data of a financial institution in a preset time period and obtains a case database, wherein the case database includes multiple financial risk cases, and M is a positive integer; matches each financial risk data with the financial risk case in the case database to obtain M groups of rank-related data, and inputs the M groups of rank-related data and the M financial risk data into a risk scoring model to output a risk score; inputs the risk score into a report generation model to output a financial risk attribution report of the financial institution, thereby solving the technical problem of low detection accuracy when performing risk detection on financial institutions in the related technology, by determining the financial risk data and the case database of a financial institution in a preset time period, matching each financial risk data with the financial risk case in the case database to obtain rank-related data, using the risk scoring model to analyze the rank-related data, outputting the risk score, and then using the report generation model to analyze the risk score to output the financial risk attribution report, thereby achieving the technical effect of significantly improving the accuracy and reliability of risk detection on financial institutions.

[0046] Optionally, in the method for determining financial risk information provided in an embodiment of the present application, determining M financial risk data of a financial institution in a preset time period includes: obtaining comprehensive risk information for the preset time period and obtaining the institution number of the financial institution; extracting financial risk information associated with the financial institution from the comprehensive risk information according to the institution number to obtain financial risk information to be processed; formatting the financial risk information to be processed to obtain converted financial risk information, and performing text segmentation on the converted financial risk information to obtain M financial risk data.

[0047] Specifically, financial risk refers to the collective opinions and expectations of market participants on the financial environment, corporate conditions or economic prospects. It can reflect the market's attitude towards uncertainty and predict potential market fluctuations and credit risks. Therefore, in order to obtain financial risk data from financial institutions, we can first obtain all public opinion information (that is, comprehensive risk information) for a certain period of time from information sources such as news reports, social media, industry analysis reports, policy statements, and economic indicators, and at the same time obtain the name or unique institutional number of the financial institution.

[0048] Furthermore, public opinion information (i.e., financial risk information) related to the institution is extracted from all public opinion information based on the above-mentioned institution number or name, thereby obtaining the financial risk information to be processed. After these are subjected to pre-processing operations such as deduplication, integration, and merging, in order to facilitate the matching of the RAG algorithm, the above-mentioned pre-processed information is vectorized and then converted into a Query format that supports large language model input, thereby obtaining the converted financial risk information. Finally, these converted financial risk information are subjected to text segmentation (Chunking), and divided into multiple smaller information units (chunks) according to semantic units, where each chunk usually includes 100-500 words, which can represent a relatively complete semantic unit, so as to facilitate subsequent retrieval and analysis, thereby obtaining multiple financial risk data.

[0049] This embodiment converts the format and segments the text of financial risk information, thereby reducing redundancy while maintaining the integrity of the information, optimizing the efficiency and accuracy of subsequent processing, and providing strong decision-making assistance for financial institutions in a complex and changing financial environment.

[0050] Optionally, in the method for determining financial risk information provided in an embodiment of the present application, each financial risk data is matched with a financial risk case in a case database to obtain M groups of rank-related data, including: extracting N financial risk cases from the case database, and vectorizing each financial risk case to obtain N candidate case data, where N is a positive integer; for one financial risk data, calculating the semantic similarity between the financial risk data and the N candidate case data respectively, to obtain N similarity data; sorting the N similarity data in descending order to obtain a similarity queue, extracting Y similarity data at a preset position from the similarity queue, and obtaining the candidate case data corresponding to the Y similarity data to obtain Y financial risk case data, where Y is less than N and Y is a positive integer; using the financial risk data and each financial risk case data for splicing to obtain Y rank-related sub-data, and combining the Y rank-related sub-data to obtain a group of rank-related data.

[0051] Specifically, after obtaining multiple financial risk data, in order to enrich the dimension and depth of public opinion analysis and ensure that the scoring results are more comprehensive and objective, we can first extract multiple cases closely related to financial risks from the case database, and then use natural language processing technology to block these cases and convert them into vector representations with unique identifiers to facilitate subsequent retrieval, thereby obtaining multiple candidate case data in the format of digital features.

[0052] Furthermore, for each piece of financial risk data to be analyzed, semantic similarity is calculated between it and all the candidate case data mentioned above. This can be measured based on the distance between vectors to obtain the corresponding similarity data. Furthermore, one or more similarity data with the highest values ​​are selected from the similarity queue formed by these similarity data. The candidate case data corresponding to these similarity data are concatenated with the financial risk data to construct a "rank-correlated sub-data." Finally, all the rank-correlated sub-data are combined to obtain a set of rank-correlated data corresponding to each piece of financial risk data.

[0053] This embodiment obtains multiple candidate case data through similarity calculation, and then obtains corresponding rank-related data by splicing the candidate case data and financial risk data. It can more carefully evaluate the positive, negative or neutral impact, avoid redundant analysis of irrelevant or low-correlation cases, greatly save computing resources and analysis time, significantly improve the accuracy, depth and practicality of risk analysis, and provide financial institutions with more solid risk management decision support.

[0054] Optionally, in the method for determining financial risk information provided in an embodiment of the present application, M groups of rank-related data and M financial risk data are input into a risk scoring model, and a risk score is output, including: adjusting the parameters of the risk scoring model according to risk identification rules to obtain an adjusted risk scoring model, wherein the risk identification rules are determined by the institutional type of the financial institution; for one financial risk data, the adjusted risk scoring model identifies the financial risk factors associated with the financial institution, a group of rank-related data, and the financial risk data to obtain K dimension scores, wherein K is a positive integer; determining the dimension weights according to the risk identification rules, and performing weighted calculation on the K dimension scores based on the dimension weights to obtain a risk score.

[0055] Specifically, after obtaining a set of rank-related data corresponding to each financial risk data set, traditional scoring models often adopt a "one-size-fits-all" strategy, ignoring the fact that different financial institutions face widely varying types and degrees of risk due to differences in their business models, market positioning, and regulatory environments. Therefore, risk identification rules can first be determined based on the institutional type of the financial institution. These rules may include factor classification, identification criteria for key risk events, and sensitivity settings for market fluctuations. Based on these rules, the parameters of the risk scoring model are then adjusted to ensure that the model can accurately identify risk factors related to the institutional type and gain a deeper understanding of the institution's specific market environment and business processes.

[0056] Furthermore, the adjusted risk scoring model is used to identify financial risk factors associated with financial institutions, a set of rank-related data, and financial risk data, and score them from multiple dimensions. For example, the emotional polarity (the positive and negative intensity of public opinion), the severity of the matter (the potential impact of public opinion events on corporate operations), and the scope of public opinion dissemination (the breadth and duration of public opinion time) are analyzed to obtain corresponding dimension scores. Finally, the above-mentioned dimension scores are weighted and summed by the determined dimension weights to obtain the risk score, which can be calculated by the following formula: S = w1 × P e +w2×I e +w3×R e Among them, S, P e , I e 、R e w1, w2, and w3 represent the weights of each dimension, respectively, representing the company's risk score, sentiment polarity score, severity score, and public opinion reach score. It's important to note that dimension weights reflect the degree to which a financial institution prioritizes different risk factors, as well as their relative importance in the overall risk assessment. For example, a financial institution heavily reliant on credit transactions could assign a higher weight to credit events.

[0057] This embodiment calculates dimensional scores from different dimensions using a risk scoring model, and obtains a final risk score by weighting these dimensional scores. This can achieve more personalized and accurate risk quantification, provide decision makers with risk insights that are closer to the actual situation of the organization, enhance the scientific nature and effectiveness of decision-making, and then formulate more comprehensive and flexible risk management strategies based on the scoring results.

[0058] Optionally, in the method for determining financial risk information provided in an embodiment of the present application, financial risk factors are determined in the following manner: obtaining the institutional type of the financial institution, determining risk identification rules based on the institutional type, wherein the risk identification rules include at least: sentiment polarity rules, operational impact rules, and propagation rules; for a financial risk data, analyzing a set of rank-related data associated with the financial risk data based on the risk identification rules to obtain financial risk factors, wherein the financial risk factors include at least risk event information and sentiment polarity associated with the risk event information.

[0059] Specifically, in order to help the risk scoring model accurately identify risk factors related to the type of institution, before inputting the financial risk data into the above model, the financial risk factors associated with each financial risk data can be first determined, and then the financial risk factors of the financial institution can be determined based on the factors. Since different types of institutions have different sensitivities and response patterns to risks due to differences in their business attributes and market positioning, risk identification rules can be customized according to the institutional type of the financial institution to more accurately capture and understand the risk factors faced by specific institutions. Among them, risk identification rules may include sentiment polarity rules, operational impact rules, and dissemination rules. For example, sentiment polarity rules focus on identifying and classifying the sentiment tendencies of data, including positive, negative, or neutral; operational impact rules involve assessing the potential impact of events on the operating conditions of financial institutions, including liquidity, asset value fluctuations, changes in customer behavior, etc.; dissemination rules focus on the impact of the speed and scope of information dissemination in the financial market on risk.

[0060] Furthermore, based on the customized risk identification rules described above, the rank-related data associated with each financial risk data set can be analyzed to determine financial risk factors that include sentiment polarity and risk event information. For example, this financial risk factor can accurately identify which events or news caused fluctuations, as well as the specific impact of these events on financial institutions (such as increased credit risk, decreased market confidence, etc.). Based on the sentiment polarity rules, the tendency in the event can be quantitatively assessed to determine whether it is biased towards negativity and its intensity. Finally, based on the financial risk factors of each financial risk data set, the overall financial risk factors of the financial institution can be determined.

[0061] This embodiment determines financial risk factors by utilizing risk identification rules, so that risk analysis is more in line with the actual needs of the institution, provides a more comprehensive perspective for risk management, and improves the scientific nature and effectiveness of decision-making.

[0062] Optionally, in the method for determining financial risk information provided in an embodiment of the present application, the risk score is input into a report generation model, and outputting a financial risk attribution report of the financial institution includes: obtaining the institutional type of the financial institution, determining a preset report template based on the institutional type; inputting the preset report template and the risk score into the report generation model, and outputting the financial risk attribution report.

[0063] Specifically, after the risk scoring model outputs a risk score, to improve the relevance and practicality of the financial risk attribution report, a pre-defined report template can be determined based on the type of financial institution. For example, for investment institutions, the template may place greater emphasis on market volatility, portfolio performance, and changes in macroeconomic indicators. This pre-defined report template and the risk score are then input into the report generation model, which then attributes and interprets the scoring process, ultimately outputting a financial risk attribution report containing detailed analysis results.

[0064] This embodiment inputs the preset report template and risk score corresponding to the financial institution into the report generation model, so that the generated report can accurately reflect the specific risk types and potential threats faced by such institutions, thereby improving the practicality and guiding value of the report as well as the efficiency and effectiveness of decision-making.

[0065] Optionally, in the method for determining financial risk information provided in an embodiment of the present application, a preset report template and a risk score are input into a report generation model, and outputting a financial risk attribution report includes: extracting K dimension scores from the risk score, wherein each dimension score is associated with a dimension type, and the dimension types include at least: sentiment polarity dimension, operational impact dimension, and communication scope dimension; obtaining dimension scoring details, analyzing the K dimension scores based on the dimension scoring details, and obtaining score explanation information associated with the risk score; inputting the score explanation information, the preset report template, and the risk score into the report generation model, and outputting a financial risk attribution report.

[0066] Specifically, in order to ensure the comprehensiveness and accuracy of risk assessment, based on the preset report template and the risk score input report generation model, it is also possible to analyze the multiple dimension scores extracted from the risk score, that is, by obtaining the dimension scoring rules, and analyzing the dimension scores based on the dimension scoring rules, for example, determining the specific proportion of positive and negative in the score, keywords and changing trends; analyzing the nature, frequency, duration of business risk events and comparative analysis with past events; evaluating the spread of public opinion in different channels, and then generating risk score-related score explanatory information based on the above analysis results.

[0067] Furthermore, the score explanation information, the preset report template and the risk score are input into the report generation model to output a financial risk attribution report.

[0068] This embodiment, by inputting data such as score explanation information, preset report templates, and risk scores into the report generation model, can enable the output report to clearly understand the composition and basis of the risk score, promote the transparency of the decision-making process, improve the scientific nature and efficiency of decision-making, and enable financial institutions to more accurately identify potential risks and take preventive measures in advance.

[0069] The present application also provides a method for determining financial risk information. Figure 3 This is an example of an optional method for determining financial risk information provided by an embodiment of the present application. Figure 1 ,like Figure 3 As shown, the method includes:

[0070] Specifically, we first obtain a series of public opinion information related to financial institutions from the enterprise public opinion database composed of multiple information sources. For example, public opinion data can include the latest market comments on financial institutions, excerpts from economic analysts' reports, and information on industry policy changes. The public opinion information is then vectorized to obtain a query format that supports large language model input. The text is then chunked and divided into multiple smaller information units (chunks) according to semantic units. Each chunk usually includes 100-500 words to facilitate subsequent retrieval and analysis. At the same time, each document in the expert knowledge case library a, which contains basic enterprise information, industry policy documents, historical public opinion data, etc., is chunked and has a unique identifier to facilitate subsequent retrieval.

[0071] Furthermore, the collected public opinion information is context-matched through the RAG algorithm, that is, the public opinion information chunk and the expert case knowledge base are spliced ​​to obtain a rank-related chunk, and then the information units of the two are vectorized and compared, their semantic similarity is calculated, and they are sorted to obtain the most relevant Top-N information units as the output of the subsequent scoring LLM.

[0072] Furthermore, the rank correlation chunk undergoes sentiment analysis and risk factor extraction, identifying the sentiment polarity (positive, neutral, negative) and intensity of the public opinion information, as well as key risk events within the public opinion. The Large Language Scoring Model (LLMa) then calculates the final public opinion score based on the public opinion information and contextual information. The Large Language Report Generation Model (LLMb) then processes the aforementioned public opinion scoring process and calculates weights, combining it with relevant contextual information from the expert case knowledge base (b) to attribute and explain the scoring process, ultimately generating an interpretable score attribution report.

[0073] This embodiment collects and analyzes public opinion data in real time, and combines the RAG algorithm and LLM model to quickly and accurately assess the public opinion risks of financial institutions, thereby helping financial institutions make more accurate and rapid decision adjustments based on the specific market environment and public opinion situation, and improve their response capabilities and competitiveness in complex market environments.

[0074] The present application also provides a system for determining financial risk information. It should be noted that the system for determining financial risk information in the present application can be used to execute the method for determining financial risk information provided in the present application. The following describes the system for determining financial risk information provided in the present application.

[0075] According to the embodiment of this application, Figure 4 is a schematic diagram of a system for determining financial risk information provided in accordance with an embodiment of the present application. Figure 4 As shown, the system includes:

[0076] The database module 401 is used to obtain M financial risk data and case databases of financial institutions in a preset time period, and send the financial risk cases in the M financial risk data and case databases to the language processing and matching module 402 and the scoring module 403;

[0077] Specifically, database module 401 is associated with a corporate public opinion database, which contains multiple financial risk data collected from various channels, such as market news, social media, industry reports, and corporate announcements, for a preset time period. To conduct risk analysis on financial institutions, multiple financial risk data and case databases for a preset time period can first be obtained from the corporate public opinion database. These data are then classified, deduplicated, and integrated before being sent to language processing and matching module 402 and scoring module 403.

[0078] The language processing and matching module 402 is used to match each financial risk data with the financial risk cases in the case database to obtain M groups of rank-related data, and send the M groups of rank-related data to the scoring module 403, where M is a positive integer;

[0079] It should be noted that the language processing matching module 402 is capable of performing information matching based on the RAG algorithm. After receiving the financial risk data and the case database, each financial risk data can be matched with the financial risk cases in the case database to obtain multiple sets of rank-related data associated with each financial risk data, and pass them to the scoring module 403. After the language processing matching module 402 receives the financial risk data, it first converts it into a semantic vector representation (that is, vectorization processing is performed) to obtain financial risk data in a query format. The financial risk data in the query format is then matched with the financial risk cases to identify the case data most relevant to the current data, and then multiple sets of rank-related data are determined based on the case data.

[0080] Scoring module 403, for inputting M sets of rank correlation data and M financial risk data into a risk scoring model, outputting a risk score, and sending the risk score to report generation module 404;

[0081] Specifically, after receiving the rank-correlation data and financial risk data, the scoring module 403 can input the rank-correlation data and financial risk data into a risk scoring model. The risk scoring model can be a large language scoring model trained using algorithms such as the Transformer architecture or a deep neural network. After receiving the rank-correlation data and financial risk data, the model extracts key risk signals from the rank-correlation data, while considering the semantics, context, and historical trends of the financial risk data. After integrating and analyzing these features, it outputs a risk score and corresponding score explanation information, which is then sent to the report generation module 404.

[0082] The report generation module 404 is used to obtain a preset report template and score explanation information associated with the risk score, input the score explanation information, the preset report template and the risk score into the report generation model, and output a financial risk attribution report.

[0083] Specifically, after receiving the risk score and the corresponding score explanation information, the report generation module 404 can input the score explanation information, the preset report template and the risk score into the report generation model after obtaining the preset report template, and output the financial risk attribution report, wherein the report generation model can refer to a large language report generation model.

[0084] The financial risk information determination system provided in the embodiment of the present application is used to obtain M financial risk data and case databases of financial institutions in a preset time period through the database module 401, and send the M financial risk data and financial risk cases in the case database to the language processing matching module 402 and the scoring module 403; the language processing matching module 402 is used to match each financial risk data with the financial risk cases in the case database to obtain M groups of rank-related data, and send the M groups of rank-related data to the scoring module 403, where M is a positive integer; the scoring module 403 is used to input the M groups of rank-related data and the M financial risk data into the risk scoring model, output a risk score, and send the risk score to the report generation module 404; the report generation module 404 is used to obtain a preset report template and risk score association The scoring explanation information, the preset report template and the risk score are input into the report generation model, and the financial risk attribution report is output, which solves the technical problem of low detection accuracy when performing risk detection on financial institutions in related technologies. Through the coordinated operation of the database module 401, the language processing matching module 402, the scoring module 403 and the report generation module 404, by determining the financial risk data and case database of the financial institution in a preset time period, each financial risk data is matched with the financial risk case in the case database to obtain rank-related data, and the rank-related data is analyzed by the risk scoring model to output the risk score. The risk score is then analyzed by the report generation model to output the financial risk attribution report, thereby achieving the technical effect of significantly improving the accuracy and reliability of risk detection on financial institutions.

[0085] The present application also provides an optional method for determining financial risk information, which is applied to Figure 4 The financial risk information determination system shown, Figure 5 This is an example of an optional method for determining financial risk information provided by an embodiment of the present application. Figure 2 ,like Figure 5 As shown, the method includes:

[0086] Specifically, in order to conduct risk analysis on financial institutions, the database module 401 can first obtain multiple financial risk data and case databases within a preset time period from the corporate public opinion database, and send them to the language processing matching module 402 and the scoring module 403 after classification, deduplication, and integration. After the language processing matching module 402 receives multiple financial risk data, it first converts them into semantic vector representations (that is, vectorized Query processing is performed) to obtain financial risk data in Query format. Each financial risk data is then matched with the financial risk cases in the case database to identify the case data most relevant to the current data, that is, to retrieve the contextual information (chunk) related to the financial institution, and then determine multiple groups of rank-related data based on the case data, and pass them to the scoring module 403.

[0087] Furthermore, after the scoring module 403 receives the rank-related data and the financial risk data, the rank-related data and the financial risk data are input into the risk scoring model. After obtaining the rank-related data and the financial risk data, the model integrates and analyzes the data, and then outputs a risk score and corresponding score explanation information, and sends it to the report generation module 404. After the report generation module 404 receives the risk score and the corresponding score explanation information, it can match the preset report template in the expert case knowledge base, and input the score explanation information, the preset report template, and the risk score into the report generation model, thereby outputting a financial risk attribution report.

[0088] This embodiment, through the coordinated operation of the database module 401, the language processing matching module 402, the scoring module 403 and the report generation module 404, determines the financial risk data and case database of the financial institution in a preset time period, matches each financial risk data with the financial risk case in the case database, obtains rank-related data, uses the risk scoring model to analyze the rank-related data, outputs the risk score, and then uses the report generation model to analyze the risk score and output the financial risk attribution report, thereby achieving the technical effect of significantly improving the accuracy and reliability of risk detection for financial institutions.

[0089] It should be noted that the steps shown in the flowcharts of the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions, and that, although a logical order is shown in the flowcharts, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0090] Example 2

[0091] The present application also provides a device for determining financial risk information. It should be noted that the device for determining financial risk information in the present application can be used to execute the method for determining financial risk information provided in the present application. The following describes the device for determining financial risk information provided in the present application.

[0092] According to an embodiment of the present application, a device for implementing the above-mentioned method for determining financial risk information is also provided. Figure 6 is a schematic diagram of a device for determining financial risk information according to an embodiment of the present application. Figure 6 As shown, the device includes: a determination unit 60, a matching unit 61, and an input unit 62.

[0093] a determination unit 60 configured to determine M financial risk data of a financial institution in a preset time period and obtain a case database, wherein the case database includes a plurality of financial risk cases, and M is a positive integer;

[0094] A matching unit 61 is configured to match each financial risk data with a financial risk case in the case database to obtain M sets of rank-related data, input the M sets of rank-related data and the M financial risk data into a risk scoring model, and output a risk score;

[0095] The input unit 62 is used to input the risk score into the report generation model and output the financial risk attribution report of the financial institution.

[0096] The financial risk information determination device provided in the embodiment of the present application determines M financial risk data of a financial institution in a preset time period through a determination unit 60, and obtains a case database, wherein the case database includes multiple financial risk cases, and M is a positive integer; the matching unit 61 matches each financial risk data with the financial risk case in the case database to obtain M groups of rank-related data, and inputs the M groups of rank-related data and the M financial risk data into a risk scoring model to output a risk score; the input unit 62 inputs the risk score into a report generation model to output a financial risk attribution report of the financial institution, thereby solving the technical problem of low detection accuracy when performing risk detection on financial institutions in related technologies, by determining the financial risk data and the case database of the financial institution in a preset time period, matching each financial risk data with the financial risk case in the case database to obtain rank-related data, using the risk scoring model to analyze the rank-related data, and then using the report generation model to analyze the risk score and output the financial risk attribution report, thereby achieving the technical effect of significantly improving the accuracy and reliability of risk detection on financial institutions.

[0097] Optionally, in the device for determining financial risk information provided in an embodiment of the present application, the determination unit 60 includes: a first acquisition module, used to obtain comprehensive risk information for a preset time period and obtain the institution number of the financial institution; a first extraction module, used to extract financial risk information associated with the financial institution from the comprehensive risk information according to the institution number to obtain financial risk information to be processed; a conversion module, used to convert the format of the financial risk information to be processed to obtain converted financial risk information, and perform text segmentation on the converted financial risk information to obtain M financial risk data.

[0098] Optionally, in the financial risk information determination device provided in the embodiment of the present application, the matching unit 61 includes: a second extraction module, used to extract N financial risk cases from the case database, and vectorize each financial risk case to obtain N candidate case data, where N is a positive integer; a calculation module, used to calculate the semantic similarity between the financial risk data and the N candidate case data for one financial risk data, and obtain N similarity data; a sorting module, used to sort the N similarity data in descending order to obtain a similarity queue, extract Y similarity data at a preset position from the similarity queue, and obtain the candidate case data corresponding to the Y similarity data to obtain Y financial risk case data, where Y is less than N and Y is a positive integer; a splicing module, used to use the financial risk data and each financial risk case data for splicing to obtain Y rank-related sub-data, and a group of rank-related data is obtained by combining the Y rank-related sub-data.

[0099] Optionally, in the device for determining financial risk information provided in an embodiment of the present application, the matching unit 61 includes: an adjustment module, used to adjust the parameters of the risk scoring model according to the risk identification rules to obtain an adjusted risk scoring model, wherein the risk identification rules are determined by the institutional type of the financial institution; an identification module, used to identify, for a piece of financial risk data, the financial risk factors associated with the financial institution, a set of rank-related data and the financial risk data by the adjusted risk scoring model to obtain K dimension scores, wherein K is a positive integer; a determination module, used to determine the dimension weights according to the risk identification rules, and perform weighted calculation on the K dimension scores based on the dimension weights to obtain a risk score.

[0100] Optionally, in the device for determining financial risk information provided in an embodiment of the present application, the matching unit 61 includes: a second acquisition module, used to obtain the institutional type of the financial institution, and determine the risk identification rules based on the institutional type, wherein the risk identification rules include at least: sentiment polarity rules, operational impact rules, and propagation rules; an analysis module, used to analyze a set of rank-related data associated with a financial risk data based on the risk identification rules to obtain financial risk factors, wherein the financial risk factors include at least risk event information and sentiment polarity associated with the risk event information.

[0101] Optionally, in the device for determining financial risk information provided in an embodiment of the present application, the input unit 62 includes: a third acquisition module, used to obtain the institutional type of the financial institution and determine the preset report template based on the institutional type; a first input module, used to input the preset report template and the risk score into the report generation model and output the financial risk attribution report.

[0102] Optionally, in the device for determining financial risk information provided in an embodiment of the present application, the input unit 62 includes: a third extraction module, used to extract K dimension scores from the risk score, wherein each dimension score is associated with a dimension type, and the dimension types include at least: sentiment polarity dimension, business impact dimension, and communication range dimension; a fourth acquisition module, used to obtain dimension scoring details, analyze the K dimension scores based on the dimension scoring details, and obtain score explanation information associated with the risk score; a second input module, used to input the score explanation information, a preset report template, and the risk score into a report generation model, and output a financial risk attribution report.

[0103] It should be noted that the above-mentioned determination unit 60, matching unit 61, and input unit 62 correspond to steps S201 to S203 in Example 1. The examples and application scenarios implemented by the above-mentioned units and corresponding steps are the same, but are not limited to the contents disclosed in the above-mentioned Example 1. It should be noted that the above-mentioned modules or units can be hardware components or software components stored in a memory (e.g., memory 104) and processed by one or more processors (e.g., processors 102a, 102b, ..., 102n). The above-mentioned units can also be part of the device and can be run in the computer terminal 10 provided in Example 1.

[0104] Example 3

[0105] The embodiment of the present application may provide a computer terminal, which may be any computer terminal device in a computer terminal group. Optionally, in this embodiment, the computer terminal may also be replaced by a terminal device such as a mobile terminal or an electronic device.

[0106] Optionally, in this embodiment, the computer terminal may be located in at least one network device among a plurality of network devices of a computer network.

[0107] In this embodiment, the above-mentioned computer terminal can execute the program code of the following steps in the method for determining financial risk information: determining M financial risk data of a financial institution in a preset time period, and obtaining a case database, wherein the case database includes multiple financial risk cases and M is a positive integer; matching each financial risk data with the financial risk case in the case database to obtain M groups of rank-related data, and inputting the M groups of rank-related data and the M financial risk data into a risk scoring model to output a risk score; inputting the risk score into a report generation model to output a financial risk attribution report of the financial institution.

[0108] Optionally, the above-mentioned computer terminal can execute the program code of the following steps in the method for determining financial risk information: obtaining comprehensive risk information for a preset time period and obtaining the institution number of the financial institution; extracting financial risk information associated with the financial institution from the comprehensive risk information according to the institution number to obtain financial risk information to be processed; converting the format of the financial risk information to be processed to obtain converted financial risk information, and performing text segmentation on the converted financial risk information to obtain M financial risk data.

[0109] Optionally, the above-mentioned computer terminal can execute the program code of the following steps in the method for determining financial risk information: extract N financial risk cases from the case database, and vectorize each financial risk case to obtain N candidate case data, where N is a positive integer; for one financial risk data, calculate the semantic similarity between the financial risk data and the N candidate case data respectively, and obtain N similarity data; sort the N similarity data in descending order to obtain a similarity queue, extract Y similarity data at a preset position from the similarity queue, and obtain the candidate case data corresponding to the Y similarity data to obtain Y financial risk case data, where Y is less than N and Y is a positive integer; use the financial risk data and each financial risk case data for splicing to obtain Y rank-related sub-data, and combine the Y rank-related sub-data to obtain a group of rank-related data.

[0110] Optionally, the above-mentioned computer terminal can execute the program code of the following steps in the method for determining financial risk information: adjusting the parameters of the risk scoring model according to the risk identification rules to obtain an adjusted risk scoring model, wherein the risk identification rules are determined by the institutional type of the financial institution; for a financial risk data, the adjusted risk scoring model identifies the financial risk factors associated with the financial institution, a set of rank-related data and the financial risk data to obtain K dimension scores, wherein K is a positive integer; determining the dimension weights according to the risk identification rules, and performing weighted calculation on the K dimension scores based on the dimension weights to obtain a risk score.

[0111] Optionally, the above-mentioned computer terminal can execute the program code of the following steps in the method for determining financial risk information: obtaining the institutional type of the financial institution, determining the risk identification rules based on the institutional type, wherein the risk identification rules include at least: sentiment polarity rules, operational impact rules, and propagation rules; for a financial risk data, analyzing a set of rank-related data associated with the financial risk data based on the risk identification rules to obtain financial risk factors, wherein the financial risk factors include at least risk event information and sentiment polarity associated with the risk event information.

[0112] Optionally, the above-mentioned computer terminal can execute the program code of the following steps in the method for determining financial risk information: obtaining the institutional type of the financial institution, determining a preset report template based on the institutional type; inputting the preset report template and risk score into the report generation model, and outputting a financial risk attribution report.

[0113] Optionally, the above-mentioned computer terminal can execute the program code of the following steps in the method for determining financial risk information: extracting K dimension scores from the risk score, wherein each dimension score is associated with a dimension type, and the dimension types include at least: sentiment polarity dimension, business impact dimension, and communication scope dimension; obtaining dimension scoring rules, analyzing the K dimension scores based on the dimension scoring rules, and obtaining score explanation information associated with the risk score; inputting the score explanation information, the preset report template, and the risk score into the report generation model, and outputting the financial risk attribution report.

[0114] Optionally, Figure 7 This is a structural block diagram of an electronic device according to an embodiment of the present application. Figure 7 As shown, the electronic device may include: one or more ( Figure 7 Only one is shown) processor 702, memory 704, storage controller, and peripheral interface, wherein the peripheral interface is connected to the radio frequency module, audio module and display.

[0115] The memory can be used to store software programs and modules, such as the program instructions / modules corresponding to the method and apparatus for determining financial risk information in the embodiments of the present application. The processor executes the software programs and modules stored in the memory to perform various functional applications and data processing, thereby implementing the above-mentioned method for determining financial risk information. 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 examples, the memory may further include memory remotely located relative to the processor, and these remote memories may be connected to the terminal via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, corporate intranets, local area networks, mobile communication networks, and combinations thereof.

[0116] The processor can call the information and application stored in the memory through the transmission device to execute the above steps in the above method for determining financial risk information.

[0117] An embodiment of the present application provides a solution for determining financial risk information. By determining M financial risk data of a financial institution in a preset time period and obtaining a case database, wherein the case database includes multiple financial risk cases, and M is a positive integer; matching each financial risk data with the financial risk cases in the case database to obtain M groups of rank-related data, and inputting the M groups of rank-related data and the M financial risk data into a risk scoring model to output a risk score; inputting the risk score into a report generation model to output a financial risk attribution report for the financial institution, the technical problem of low detection accuracy when performing risk detection on financial institutions in related technologies is solved. By determining the financial risk data of a financial institution in a preset time period and a case database, matching each financial risk data with the financial risk cases in the case database to obtain rank-related data, using the risk scoring model to analyze the rank-related data and output a risk score, and then using the report generation model to analyze the risk score and output a financial risk attribution report, the technical effect of significantly improving the accuracy and reliability of risk detection on financial institutions is achieved.

[0118] It can be understood by those skilled in the art that Figure 7 The structure shown is for illustration only, and the electronic device may also be a terminal device such as a smart phone, a tablet computer, a PDA, a mobile Internet device (MID), or a PAD. Figure 7 It does not limit the structure of the above electronic device. For example, the electronic device may also include Figure 7 More or fewer components (such as network interfaces, display devices, etc.) shown in, or with Figure 7 Different configurations shown.

[0119] A person skilled in the art will understand that all or part of the steps in the various methods of the above embodiments can be completed by instructing the hardware related to the terminal device through a program, and the program can be stored in a computer-readable storage medium, which may include: a flash drive, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, etc.

[0120] Example 4

[0121] The embodiment of the present application further provides a storage medium. Optionally, in this embodiment, the storage medium can be used to store the program code executed by the method for determining financial risk information provided in the first embodiment.

[0122] Optionally, in this embodiment, the storage medium may be located in any computer terminal in a computer terminal group in a computer network, or in any mobile terminal in a mobile terminal group.

[0123] Optionally, in this embodiment, the storage medium is configured to store program code for executing the following steps: determining M financial risk data of a financial institution in a preset time period, and obtaining a case database, wherein the case database includes multiple financial risk cases and M is a positive integer; matching each financial risk data with the financial risk case in the case database to obtain M groups of rank-related data, and inputting the M groups of rank-related data and the M financial risk data into a risk scoring model to output a risk score; inputting the risk score into a report generation model to output a financial risk attribution report of the financial institution.

[0124] The present application also provides a computer program product, which, when executed on a data processing device, is suitable for executing the steps of the method for determining financial risk information.

[0125] The serial numbers of the above embodiments of the present application are for description only and do not represent the advantages or disadvantages of the embodiments.

[0126] In the above embodiments of the present application, the description of each embodiment has its own focus. For parts that are not described in detail in a certain embodiment, please refer to the relevant description of other embodiments.

[0127] In the several embodiments provided in this application, it should be understood that the disclosed technical content can be implemented in other ways. Among them, the device embodiments described above are only schematic. For example, the division of units is only a logical function division. In actual implementation, there may be other division methods. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of units or modules, which can be electrical or other forms.

[0128] Units described as separate components may or may not be physically separate, and 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 these units may be selected to achieve the purpose of this embodiment according to actual needs.

[0129] In addition, the functional units in the various embodiments of the present application may be integrated into a single processing unit, or each unit may exist physically separately, or two or more units may be integrated into a single unit. The aforementioned integrated units may be implemented in the form of hardware or software functional units.

[0130] If the integrated unit is implemented in the form of 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 the present application is essentially 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, and the computer software product is stored in a storage medium, including a number of instructions for enabling a computer device (which can be a personal computer, server or network device, etc.) to execute all or part of the steps of the various embodiments of the present application. The aforementioned storage medium includes: U disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), mobile hard disk, magnetic disk or optical disk and other media that can store program code.

[0131] The above is only a preferred embodiment of the present application. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present application. These improvements and modifications should also be regarded as the scope of protection of the present application.

Claims

1. A method for determining financial risk information, characterized in that: include: Determine M financial risk data of a financial institution in a preset time period, and obtain a case database, wherein the case database includes multiple financial risk cases, and M is a positive integer; Matching each financial risk data with the financial risk cases in the case database to obtain M groups of rank-related data, and inputting the M groups of rank-related data and the M financial risk data into a risk scoring model to output a risk score; The risk score is input into a report generation model to output a financial risk attribution report of the financial institution.

2. The method according to claim 1, characterized in that Determine the M financial risk data of a financial institution in a preset time period, including: Obtaining comprehensive risk information for the preset time period and obtaining the institution number of the financial institution; extracting financial risk information associated with the financial institution from the comprehensive risk information according to the institution number to obtain financial risk information to be processed; The format of the financial risk information to be processed is converted to obtain converted financial risk information, and the converted financial risk information is text-segmented to obtain the M financial risk data.

3. The method according to claim 1, characterized in that Each financial risk data is matched with the financial risk cases in the case database to obtain M groups of rank-related data including: Extracting N financial risk cases from the case database, and performing vectorization processing on each financial risk case to obtain N candidate case data, where N is a positive integer; For a piece of financial risk data, calculating the semantic similarity between the financial risk data and the N candidate case data respectively to obtain N similarity data; Sorting the N similarity data in descending order to obtain a similarity queue, extracting Y similarity data at a preset position from the similarity queue, and obtaining candidate case data corresponding to the Y similarity data to obtain Y financial risk case data, where Y is less than N and Y is a positive integer; The financial risk data and each financial risk case data are spliced ​​together to obtain Y rank-correlated sub-data, and the Y rank-correlated sub-data are combined to obtain a set of rank-correlated data.

4. The method according to claim 1, wherein Inputting the M groups of rank-related data and the M financial risk data into a risk scoring model and outputting a risk score includes: Adjusting parameters of the risk scoring model according to risk identification rules to obtain an adjusted risk scoring model, wherein the risk identification rules are determined by the type of the financial institution; For a piece of financial risk data, the adjusted risk scoring model identifies the financial risk factors associated with the financial institution, a set of rank-related data, and the financial risk data to obtain K dimension scores, where K is a positive integer; The dimension weights are determined according to the risk identification rules, and the K dimension scores are weighted calculated based on the dimension weights to obtain the risk score.

5. The method according to claim 4, characterized in that The financial risk factors are determined in the following manner: Obtaining the institution type of the financial institution, and determining risk identification rules based on the institution type, wherein the risk identification rules include at least: a sentiment polarity rule, an operational impact rule, and a propagation rule; For a piece of financial risk data, a group of rank-related data associated with the financial risk data is analyzed based on the risk identification rule to obtain a financial risk factor, wherein the financial risk factor includes at least risk event information and sentiment polarity associated with the risk event information.

6. The method according to claim 1, characterized in that Inputting the risk score into a report generation model and outputting a financial risk attribution report of the financial institution includes: Obtaining the institution type of the financial institution, and determining a preset report template based on the institution type; The preset report template and the risk score are input into the report generation model to output the financial risk attribution report.

7. The method according to claim 6, characterized in that Inputting the preset report template and the risk score into the report generation model and outputting the financial risk attribution report includes: Extracting K dimension scores from the risk score, wherein each dimension score is associated with a dimension type, and the dimension types include at least: a sentiment polarity dimension, a business impact dimension, and a communication scope dimension; Obtaining dimension scoring details, analyzing the K dimension scores based on the dimension scoring details, and obtaining score explanation information associated with the risk score; The score explanation information, the preset report template and the risk score are input into the report generation model, and the financial risk attribution report is output.

8. A system for determining financial risk information, characterized in that: include: A database module is used to obtain M financial risk data and a case database of a financial institution in a preset time period, and send the M financial risk data and the financial risk cases in the case database to the language processing matching module and the scoring module; The language processing and matching module is configured to match each financial risk data with the financial risk cases in the case database to obtain M groups of rank-related data, and send the M groups of rank-related data to the scoring module, where M is a positive integer; The scoring module is configured to input the M groups of rank-related data and the M financial risk data into a risk scoring model, output a risk score, and send the risk score to a report generation module; The report generation module is used to obtain a preset report template and score explanation information associated with the risk score, input the score explanation information, the preset report template and the risk score into a report generation model, and output a financial risk attribution report.

9. An electronic device, characterized in that: include: a memory storing an executable program; A processor is used to run the program, wherein when the program is run, the method for determining financial risk information according to any one of claims 1 to 7 is executed.

10. A computer program product comprising computer instructions, characterized in that When the computer instructions are executed by a processor, the steps of the method for determining financial risk information according to any one of claims 1 to 7 are implemented.

Images