Enterprise management level risk dynamic evaluation method and system

Abstract

The invention provides an enterprise management level risk dynamic evaluation method and system, relates to the technical field of computer data analysis, and solves the problem of complex risk evaluation requirements in enterprise management. The method comprises the steps of collecting data, related to enterprise risk management, of a web portal through a third-party interface, and calling enterprise system data from a local system; querying enterprise risk feature information worthy of attention at the current time from data of a third-party web portal in a structured query mode; classifying the feature information to obtain enterprise risk categories; extracting related information from the enterprise system data according to the enterprise risk category; analyzing the analysis data, and calculating the risk value of each enterprise risk category according to the occurrence frequency and quantity of the data under each enterprise risk category.

Description

Technical Field

[0001] This application relates to the technical field of computer data analysis, and particularly to a method and system for dynamically evaluating the risk of enterprise management level. Background Art

[0002] Data analysis in the big data era has become the core tool for solving complex problems. Traditional statistical methods are difficult to meet the analysis requirements of complex problems and need to rely on professional programs and equipment for in-depth statistical processing. In enterprise management, how to extract value from massive information and conduct comprehensive comparisons is a major challenge. Therefore, it is necessary to develop a program system that can quickly, accurately, and comprehensively evaluate and analyze the risk of enterprise management level.

[0003] The main disadvantages of the current enterprise management level risk assessment system include: In terms of data accuracy, there may be incomplete or incorrect information, affecting the reliability of the evaluation results. In terms of algorithm models, there are situations of being outdated or too simple, making it difficult to meet the complex risk assessment requirements in enterprise management. In terms of result interpretation, it is not clear and definite enough, making it difficult to provide specific and effective improvement suggestions for enterprises based on the evaluation results. Summary of the Invention

[0004] In view of the above-mentioned disadvantages of the related technologies, this application provides a method and system for dynamically evaluating the risk of enterprise management level to solve the above technical problems.

[0005] In a first aspect, a method for dynamically evaluating the risk of enterprise management level provided by this application includes: Collect data related to enterprise risk management on the portal website through a third-party interface, and retrieve enterprise system data from the local system; Query the enterprise risk characteristic information worthy of attention at the current time from the data of the third-party portal website in a structured query manner; Classify the characteristic information to obtain enterprise risk categories; Extract relevant information from the enterprise system data according to the enterprise risk categories; Analyze the analysis data, and calculate the risk values of each enterprise risk category according to the occurrence frequency and quantity of data under each enterprise risk category.

[0006] In an embodiment of this application, classifying the characteristic information to obtain enterprise risk categories includes: Cluster the characteristic information to obtain multiple clusters; Input the characteristics in each cluster into a pre-established decision tree model respectively, and output the enterprise risk categories of each cluster.

[0007] In an embodiment of this application, the method further includes: Determine whether the risk value exceeds a specific threshold. If the risk value of the risk category of the target enterprise exceeds the threshold, return this category to the feature extraction unit; After the feature extraction unit queries the enterprise risk feature information worthy of attention at the current time next time, supplement the risk feature information by querying the keywords under the human resource risk.

[0008] In one embodiment of the present application, the method further includes: Generate a risk heat map and display the risk heat map through a risk dashboard.

[0009] In one embodiment of the present application, the method further includes: Perform data quality monitoring on the data collected from each platform, and perform model performance evaluation on the process of the model calculation unit calculating the risk classification result to verify the confidence level of the risk classification result; Extract relevant information from the enterprise system data according to the enterprise risk category, including: If the confidence level of verifying the risk classification result is higher than the confidence level threshold, extract relevant information from the enterprise system data according to the enterprise risk category.

[0010] In one embodiment of the present application, before querying the enterprise risk feature information worthy of attention at the current time from the data of the third-party portal website by means of structured query, the method further includes: Retrieve the data collected from the enterprise system, analyze to obtain the industry information of the enterprise, and find the enterprise risk timeliness keyword with the highest occurrence probability in the data of the data warehouse; Calculate the similarity between the enterprise risk timeliness keyword and the industry information of the enterprise; If the similarity is lower than the similarity threshold, delete the data information related to the enterprise risk timeliness keyword in the data warehouse.

[0011] In one embodiment of the present application, extract relevant information from the enterprise system data according to the enterprise risk category, including: Extract the keywords of the enterprise risk category; Load the NLP natural language library and identify the variant forms or synonyms of the keywords from the natural language library; Use regular expressions to extract the information including the keywords, variant forms of the keywords, and synonyms of the keywords in the enterprise system data.

[0012] In a second aspect, an enterprise management level risk dynamic evaluation system provided by the present application includes: Data interface, data standardization module, data model construction module, analysis engine module, data warehouse; the data standardization module further includes a data processing unit and a data management unit. The data model construction module further includes a feature extraction unit and a model calculation unit.

[0013] In an embodiment of the present application, the data interface is used to obtain information data from multiple data sources and store it in the data warehouse.

[0014] The data standardization module is used to preprocess the data; The data model construction module is used to query enterprise risk-related data, then extract features from the risk-related data, and calculate the extracted features to obtain enterprise risk categories; The analysis engine module is used to extract data features related to enterprise risk categories from the data collected from the enterprise system, and conduct risk assessment based on the data features related to enterprise risk categories.

[0015] As described above, the enterprise management level risk dynamic assessment method and system provided by the present application have the following beneficial effects: With the help of an advanced big data platform, widely collect internal and external data of the enterprise, enrich the analysis data dimension, improve the accuracy and real-time of analysis, enhance the comprehensiveness and effectiveness of risk assessment; through precise data standardization and deduplication processing, ensure data quality, efficiently generate an enterprise management level assessment analysis report, and achieve fast and accurate assessment.

[0016] The present invention obtains data from third-party portal websites, queries enterprise risk feature information that is more worthy of attention at the current time from the data of third-party portal websites through a structured query method, classifies enterprise risks based on the queried enterprise risk feature information, extracts information data related to enterprise risk classification from the enterprise management system based on the enterprise risk classification result, and analyzes the information data through the analysis engine to evaluate the score of enterprise risk classification. The above process introduces external environment data of the enterprise to establish a benchmark for risk assessment, enriches the analysis data dimension, improves the accuracy and real-time of analysis, enhances the comprehensiveness and effectiveness of risk assessment, better adapts to changes in the external environment, and provides strong data support for enterprise decision-making. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] The accompanying drawings here are incorporated into the specification and form a part of this specification, showing embodiments consistent with the present application, and are used together with the specification to explain the principles of the present application. Obviously, the accompanying drawings in the following description are only some embodiments of the present application. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts. In the drawings: Figure 1It is a schematic flowchart of the method for dynamically evaluating the risk of enterprise management level proposed in the embodiments of the present invention; Figure 2 It is a schematic structural diagram of the enterprise management level risk dynamic evaluation system; Figure 3 It is a risk heat map in an example shown by the risk instrument; Figure 4 It is a step flowchart of the method for dynamically evaluating the risk of enterprise management level proposed in the embodiments of the present invention. Specific embodiments

[0018] The following will illustrate the embodiments of the present application with reference to the accompanying drawings and preferred embodiments. Those skilled in the art can easily understand other advantages and effects of the present application from the content disclosed in this specification. The present application can also be implemented or applied through other different specific embodiments. Various details in this specification can also be modified or changed based on different viewpoints and applications without departing from the spirit of the present application. It should be understood that the preferred embodiments are only for illustrating the present application, rather than for limiting the protection scope of the present application.

[0019] It should be noted that the drawings provided in the following embodiments only illustrate the basic concept of the present application in a schematic manner. Therefore, only the components related to the present application are shown in the drawings, rather than being drawn according to the number, shape, and size of the components in actual implementation. The type, quantity, and ratio of each component in actual implementation can be arbitrarily changed, and the component layout type may also be more complex.

[0020] In the following description, a large number of details are discussed to provide a more thorough explanation of the embodiments of the present application. However, it is obvious to those skilled in the art that the embodiments of the present application can be implemented without these specific details. In other embodiments, well-known structures and devices are shown in the form of block diagrams rather than in detail to avoid making the embodiments of the present application difficult to understand.

[0021] Figure 1 It is a schematic flowchart of the method for dynamically evaluating the risk of enterprise management level proposed in the embodiments of the present invention, as Figure 1As shown in the figure, the key to realizing the dynamic evaluation of enterprise management level risks in the present invention is to obtain data from third-party portals through methods such as API interfaces, RSS subscriptions, public databases, social media, data trading platforms, web crawler frameworks, blockchain technology, and cloud computing services. By means of structured queries, enterprise risk characteristic information that is more worthy of attention at the current time is queried from the data of third-party portals. Based on the queried enterprise risk characteristic information, enterprise risk classification is carried out. Based on the enterprise risk classification results, information data related to enterprise risk classification is extracted from the enterprise management system. The information data is analyzed by an analysis engine to evaluate the score of enterprise risk classification. The above process introduces external environment data of the enterprise to establish a benchmark for risk assessment, enriches the analysis data dimension, improves the accuracy and real-time performance of the analysis, enhances the comprehensiveness and effectiveness of risk assessment, better adapts to changes in the external environment, and provides strong data support for enterprise decision-making.

[0022] The sections of the enterprise management system can include the human resources section, business management section, intellectual property protection section, customer service section, supply chain management section, strategy section, finance section, production management section, etc. Which information to extract from which section is determined by the enterprise risk classification results.

[0023] The method for dynamically evaluating enterprise management level risks proposed in the embodiment of the present invention is applied to the system for dynamically evaluating enterprise management level risks proposed in the embodiment of the present invention. Figure 2 It is a schematic diagram of the structure of the system for dynamically evaluating enterprise management level risks, as Figure 2 shown. The system includes: a data interface 201, a data standardization module 202, a data model construction module 203, an analysis engine module 204, and a data warehouse 205. The data standardization module 202 further includes a data processing unit 2021 and a data management unit 2022. The data model construction module 203 further includes a feature extraction unit 2031 and a model calculation unit 2032.

[0024] When the system for dynamically evaluating enterprise management level risks executes the method for dynamically evaluating enterprise management level risks, the operation processes of each module can include: The data interface 201 is called to obtain information data from multiple data sources and store it in the data warehouse. For example, news and social media, industry data analysis, network public data, and public notice systems can all be used as data sources. The data warehouse also retrieves data from the enterprise's internal ERP system and stores it in the data warehouse.

[0025] The data standardization module is used to preprocess data. The content of the preprocessing includes: (1) data cleaning, such as handling missing values, outliers, deduplication operations, etc.; (2) standardizing the data structure, such as converting unstructured data such as JSON and XML into structured tables and adjusting the fields extracted from different data sources to be consistent; (3) handling missing values, for example, unifying the missing value marking methods of different data sources and filling the missing values of all data according to the business logic according to the missing value marking method.

[0026] In particular, during the data cleaning process, the data collected from the enterprise ERP system is retrieved, the industry information of the enterprise is analyzed, the enterprise risk time-limited keywords with the highest occurrence probability in the data warehouse are found, and the similarity between the enterprise risk time-limited keywords and the industry information of the enterprise is calculated. If the similarity is low, the data information related to the enterprise risk time-limited keywords is deleted from the data warehouse. The deletion standard can be the average value of other risk data.

[0027] Methods such as distance and cosine similarity of the word vector word2vec are used to calculate the similarity between the enterprise risk time-limited keywords and the industry information of the enterprise.

[0028] For example, during a current period of time, due to certain reasons, there is a lot of information related to the first category of risks in the network database. If this enterprise belongs to the second industry and there is no first category of risks, if the data collected from each data source is directly used for the analysis of which enterprise risk types are worthy of attention in the current environment, the information related to the first category of risks will cause data deviation. Therefore, it is necessary to delete the information related to the first category of risks from the data collected from each data source to ensure that there is no data with particularly obvious timeliness in the data warehouse.

[0029] The data standardization module 202 completes the preprocessing of data. The data model construction module 203 is used to execute the following process: query the enterprise risk-related data through the SQL structured language, then extract the features of these risk-related data, calculate the extracted features, and then obtain the risk classification worthy of evaluation in the current environment. The specific feature extraction unit 2031 is used for feature extraction, and the model calculation unit 2032 is used to calculate the extracted features.

[0030] The classification method can adopt a clustering algorithm or combine a decision tree to calculate the features and obtain the enterprise risk classification.

[0031] The data management unit 2022 of the data standardization module 202 is used to monitor the data quality of the data collected from each platform and evaluate the model performance during the process of the model calculation unit calculating the risk classification result, so as to verify the confidence level of the risk classification result.

[0032] If the confidence level of the risk classification result verified by the data management unit 2022 is higher than the confidence level threshold, the analysis engine module 204 is used to extract data features related to the risk classification result, that is, related to the enterprise risk category, from the data collected from the enterprise ERP system, perform a risk assessment based on the data features related to the risk classification result, and output the risk assessment to the risk dashboard.

[0033] The dashboard displays a risk heat map.

[0034] The result of the risk assessment by the analysis engine module 204 or the data calculated during the process can also be input back into the model calculation module in reverse to provide a reference for weight adjustment in the model calculation module. For example, if the analysis engine module 204 calculates each data feature and outputs a risk heat map in an example shown in Figure 3 the risk dashboard, not only visualizes the risks, such as using a risk heat map to display the risk distribution, enabling enterprise managers to more intuitively understand the risk situation, obtaining that there are relatively large risks in the enterprise's human resources, but also generates a weight adjustment instruction for human resources and sends it to the feature extraction unit, and the feature extraction unit adjusts the weight of the corresponding information of the human resources risk during the next judgment process of the enterprise risk worthy of attention at present.

[0035] Figure 4 is the step flow chart of the enterprise management level risk dynamic assessment method proposed in the embodiment of the present invention. As shown in Figure 4 shown, the steps of the enterprise management level risk dynamic assessment method include: S001: Collect data related to enterprise risk management on the portal website through a third-party interface, and retrieve enterprise system data from the local system.

[0036] Pre-determine the scope where the data related to enterprise risk management is located. After collecting the data, give positive feedback according to the result of each data analysis, and adaptively adjust the data collection scope through iteration. The pre-determined scope where the data related to enterprise risk management is located involves: strategic planning, corporate governance, social responsibility, human resources, business operation risk, intellectual property protection, operation efficiency, risk management, etc.

[0037] S002: Query the enterprise risk characteristic information worthy of attention at the current time from the data of the third-party portal website in a structured query manner.

[0038] S003: Classify the characteristic information to obtain the enterprise risk category.

[0039] Collect the external environment data related to enterprise risk at the current time, perform classification calculation on the external environment data, and obtain the risk categories that should be generally noted in the external environment of the enterprise at the current time.

[0040] Before executing S002, step S012 can be executed to perform data preprocessing on the data of the third-party portal website. In another implementation, a data warehouse can also be constructed to store the data of the third-party portal website. The data warehouse can also store the enterprise system data collected from the enterprise ERP system. Then, execute S012 to perform data preprocessing on the information in the data warehouse uniformly.

[0041] S012: Perform data preprocessing on the collected data.

[0042] It includes data cleaning to remove invalid, incorrect, or duplicate data; data standardization to make the data from different sources comparable; through these operations, the data quality is improved, laying a foundation for accurately assessing risks subsequently.

[0043] On the other hand, the specific process of executing step S012 for data preprocessing is as follows: perform data preprocessing on the collected data, retrieve the data collected from the enterprise ERP system, analyze to obtain the industry information of the enterprise, search for the enterprise risk time-effective keywords with the highest occurrence probability in the data of the data warehouse, calculate the similarity between the enterprise risk time-effective keywords and the industry information of the enterprise. If the similarity is low, delete the data information related to the enterprise risk time-effective keywords in the data warehouse.

[0044] The specific way to execute S003 to output the enterprise risk category can be: S301: Cluster the feature information to obtain multiple clusters; S302: Input the features in each cluster into the pre-established decision tree model respectively, and output the enterprise risk category of each cluster.

[0045] The clustering algorithm itself belongs to unsupervised learning, which divides the feature information through the internal logical connection of the information. To ensure the accuracy of the clustering result, the method of calculating the silhouette coefficient can be adopted to verify the rationality of the clustering.

[0046] For a single sample , belonging to the cluster ; (1) Calculate its average distance to each cluster, and take the minimum value of the average distance .

[0047] = (1); is other clusters except , j is the sample in , The Euclidean distance is adopted.

[0048] (2) Calculate its average distance to other samples in the same cluster .

[0049] = (2); is the number of samples.

[0050] (3) Calculate the silhouette coefficient of the sample ; ; = (3); (4) Calculate the overall silhouette coefficient ; = (4); N is the total number of samples.

[0051] Judge the clustering effect according to the value range in which the silhouette coefficient falls. For example, a silhouette coefficient between 0.7 and 1.0 indicates an excellent clustering effect.

[0052] It is possible to collect various data information related to the corporate trend as training samples.

[0053] For example, the following risk types are preset in advance: human resource risk, business operation risk, legal risk, intellectual property risk, operation efficiency risk, risk management risk, marketing risk, etc.

[0054] For each type of risk, the following operations can be performed: Collect data information related to the risk, extract features from the collected data information, use the risk type as a label, and train the decision tree model to select different leaf nodes based on the features until the decision tree model passes through each layer of leaf nodes based on the features and accurately selects the decision route based on the same type of three-dimensional feature information, and outputs the prediction category that is the same as the label.

[0055] Taking legal risk as an example, the collected legal review records, the number of litigation cases, the contracts signed by the enterprise, the contract dispute incidence rate, the applications and maintenance of the enterprise's patents, trademarks, copyrights and other intellectual properties, the compliance review results, etc. can all be used as the training data for the legal risk category, and the above data are labeled with the label of legal risk.

[0056] S004: Extract relevant information from the enterprise system data according to the enterprise risk category.

[0057] Specifically, after keyword expansion, relevant analysis data related to each enterprise risk category can be extracted from the enterprise system data by means of keyword matching, fuzzy matching, natural language NLP entity recognition, etc.

[0058] S004 includes sub-steps: S401: Extract keywords for enterprise risk categories; For example, if the enterprise risk category is human resource risk, then extract the keywords: human, human resources.

[0059] S402: Load the NLP natural language library and identify variant forms or synonyms of the keywords from the natural language library; Identify variant forms or synonyms of the keywords from the natural language library: HR, Human Resources Department, Talent Development, Talent Development Department, Human Resources Center.

[0060] S403: Use regular expressions to extract information containing the keywords, variant forms of the keywords, and synonyms of the keywords from the enterprise system data.

[0061] For example, query the data of the enterprise system, match the information on employee departures, which can be extracted under the category of human resource risk, and the records of departure applications can be extracted under the category of human resource risk. Match the records of contract returns, which can be extracted under the category of sales risk or customer management risk.

[0062] S005: Analyze the analysis data, and calculate the risk values of each enterprise risk category according to the occurrence frequency and quantity of the data under each category.

[0063] For example, query the data of the enterprise system during a certain period, match that the frequency of information on employee departures is high, the number of departing employees is large, and the departure applications involve sensitive words. Through the analysis engine, the risk score of human resource risk is calculated to be 4.8. Query the data of the enterprise system during the same period, match that the data on the on-time delivery rate of suppliers shows that the on-time delivery rate of suppliers is high, the suppliers in the supply chain are stable, the average running time of production equipment and the product production time are within a certain range. Through the analysis engine, the score of business operation risk is calculated to be 2.0. The range of the risk score is [0, 5], and the higher the score, the greater the corresponding enterprise risk.

[0064] In addition to calculating the risk values of each enterprise risk category, the analysis engine will also execute S006: Determine whether the risk value exceeds a pre-set specific threshold. If the risk value of the target enterprise risk category exceeds the threshold, return this category to the feature extraction unit, so that when the feature extraction unit executes step S002 next time, after querying the enterprise risk feature information worthy of attention at the current time value, supplement the query of risk feature information for the target enterprise risk category.

[0065] For example, when the analysis engine specifically executes an example, it sequentially executes S006 for each enterprise risk category. During the process of determining whether the risk value exceeds a pre-set specific threshold by executing S006, when executing S006 for the human resources risk category, if it is determined that the risk value of the human resources risk category exceeds the specific threshold, during the next calculation of the risk category, after the feature extraction unit queries the enterprise risk feature information worthy of attention at the current time, it will supplement the risk feature information for the keywords under the human resources risk category again to increase the weight of the human resources risk in the next risk assessment. Even if the external environment no longer pays attention to the human resources risk next time, the system will still evaluate the enterprise risk category of human resources to complete the tracking of the section with greater risk in the previous time. On the other hand, it also integrates internal and external data to establish a risk assessment index system and a scoring model that can adapt to environmental changes.

[0066] The specific threshold can be obtained by pre-setting.

[0067] For the enterprise risk category with a relatively high risk category in the previous time, to supplement the risk feature information for the keywords, the information related to the keywords can be directly queried in the historical query.

[0068] For example, if the risk value of the human resources risk category is higher than the threshold, the human resources risk category is returned to the feature extraction unit. The feature extraction unit extracts the keyword "human resources" and queries the information related to "human resources" in the data records of the third-party portal website extracted historically and supplements it to the enterprise risk feature information.

[0069] The analysis engine of the enterprise management level risk dynamic assessment system proposed by the present invention is also connected to a risk dashboard. The risk dashboard displays data in the form of images or tables. Therefore, the enterprise management level risk dynamic assessment method further includes S007: generating a risk heat map and displaying the risk heat map through the risk dashboard.

[0070] In an example of the present invention, the analysis engine can also combine the data obtained from the third-party portal website, query the processing strategy for specific enterprise risks from the data of the third-party portal website, and generate suggestions and analyses based on the processing strategy.

[0071] The data management unit of the enterprise management level risk dynamic assessment system proposed by the present invention will also execute S008: perform data quality monitoring on the data collected from each platform, and perform model performance evaluation on the process of the model calculation unit calculating the risk classification result, so as to verify the confidence level of the risk classification result.

[0072] Embodiments of the present application further provide an electronic device, including: one or more processors; a storage device for storing one or more programs, which, when executed by the one or more processors, cause the electronic device to implement the enterprise management level risk dynamic assessment method provided in each of the above embodiments.

[0073] Another aspect of the present application further provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor of a computer, the computer is caused to execute the enterprise management level risk dynamic assessment method provided in each of the above embodiments. The computer-readable storage medium may be included in the electronic device described in the above embodiments, or may exist alone without being assembled into the electronic device.

[0074] Another aspect of the present application further provides a computer program product or a computer program. The computer program product or the computer program includes computer instructions, and the computer instructions are stored in a computer-readable storage medium. A processor of a computer device reads the computer instructions from the computer-readable storage medium, and the processor executes the computer instructions, causing the computer device to execute the enterprise management level risk dynamic assessment method provided in each of the above embodiments.

[0075] In the embodiments of the present application, the terms "first" and "second" are only used for descriptive purposes and cannot be construed as indicating or implying relative importance. The terms "comprising" and "including" mentioned throughout the specification and claims are open-ended terms and should be interpreted as "including but not limited to".

[0076] The above embodiments are only used to exemplarily illustrate the principles and effects of the present application, rather than to limit the present application. Any person familiar with this technology can modify or change the above embodiments without departing from the spirit and scope of the present application. Therefore, all equivalent modifications or changes completed by those with ordinary knowledge in the technical field without departing from the spirit and technical idea disclosed in the present application should still be covered by the claims of the present application.

Claims

1. A dynamic risk assessment method for enterprise management level, characterized in that The method includes: Collect data related to enterprise risk management from the portal website through a third-party interface, and retrieve enterprise system data from the local system; Query the enterprise risk characteristic information worthy of attention at the current time from the data of the third-party portal website through a structured query method; Classify the characteristic information to obtain enterprise risk categories; Extract relevant information from the enterprise system data according to the enterprise risk categories; Analyze the analysis data, and calculate the risk values of each enterprise risk category according to the occurrence frequency and quantity of data under each enterprise risk category.

2. The method according to claim 1, wherein Classify the characteristic information to obtain enterprise risk categories, including: Cluster the characteristic information to obtain multiple clusters; Input the characteristics in each cluster into a pre-established decision tree model respectively, and output the enterprise risk category of each cluster.

3. The method according to claim 1, characterized in that, The method further includes: Judge whether the risk value exceeds a specific threshold. If the risk value of the target enterprise risk category exceeds the threshold, return this category to the feature extraction unit; After the feature extraction unit queries the enterprise risk characteristic information worthy of attention at the current time next time, query supplementary risk characteristic information for the keywords under the target enterprise risk category.

4. The method according to claim 1, wherein The method further includes: Generate a risk heat map, and display the risk heat map through a risk dashboard.

5. The method according to claim 1, wherein The method further includes: Perform data quality monitoring on the data collected from each platform, and perform model performance evaluation on the process of the model calculation unit calculating the risk classification result to verify the confidence level of the risk classification result; Extract relevant information from the enterprise system data according to the enterprise risk categories, including: If the confidence level of verifying the risk classification result is higher than the confidence level threshold, extract relevant information from the enterprise system data according to the enterprise risk categories.

6. The method according to claim 1, wherein Before querying the enterprise risk characteristic information worthy of attention at the current time from the data of the third-party portal website through a structured query method, the method further includes: Retrieve the data collected from the enterprise system, analyze to obtain the industry information of the enterprise, and find the enterprise risk time-limit keyword with the highest occurrence probability in the data of the data warehouse; Calculate the similarity between the enterprise risk time-limit keyword and the industry information of the enterprise; If the similarity is lower than the similarity threshold, delete the data information related to the enterprise risk time-limit keyword in the data warehouse.

7. The method according to claim 1, wherein Extract relevant information from the enterprise system data according to the enterprise risk categories, including: Extract the keywords of the enterprise risk category; Load the NLP natural language library, and identify the variant forms or synonyms of the keywords from the natural language library; Use regular expressions to extract the information containing the keywords, the variant forms of the keywords, and the synonyms of the keywords from the enterprise system data.

8. A dynamic risk assessment system for enterprise management level, characterized in that, The system includes: A data interface, a data standardization module, a data model construction module, an analysis engine module, and a data warehouse; the data standardization module further includes a data processing unit and a data management unit; The data model construction module further includes a feature extraction unit and a model calculation unit.

9. The system according to claim 8, wherein The data interface is used to obtain information data from multiple data sources and store them in the data warehouse; The data standardization module is used to preprocess the data; The data model construction module is used to query enterprise risk-related data, extract features from the risk-related data, calculate the extracted features, and obtain the enterprise risk categories; The analysis engine module is used to extract data features related to enterprise risk categories from the data collected from the enterprise system, and conduct risk assessment based on the data features related to enterprise risk categories.

Images