An enterprise risk analysis method, device, storage medium and program product

Abstract

The application provides an enterprise risk analysis method, device, storage medium and program product, wherein the enterprise risk analysis method comprises the following steps: obtaining enterprise risk data of a target enterprise, wherein the enterprise risk data comprises offline government-enterprise data and specific data, and personnel data collected by a face recognition device in real time; inputting the government-enterprise data, the specific data and the personnel data into an enterprise risk early warning model to determine enterprise risk points of the target enterprise, wherein the enterprise risk points are used to indicate an enterprise risk level of the target enterprise. The application is used for improving the evaluation accuracy of enterprise risk.

Description

Technical Field

[0001] This invention relates to the field of big data analytics, and in particular to a method, apparatus, storage medium, and program product for analyzing enterprise risk. Background Technology

[0002] Currently, enterprises face increasingly fierce market competition and unprecedented challenges. Enhancing corporate risk awareness, implementing comprehensive risk management, and strengthening risk assessment are crucial means for enterprises to improve their operational capabilities. At the same time, risk monitoring plays a vital role in enterprise operations.

[0003] How to conduct enterprise risk assessment has become an urgent technical problem to be solved. Summary of the Invention

[0004] This invention provides a method, apparatus, storage medium, and program product for analyzing enterprise risks, which can improve the accuracy of enterprise risk assessment.

[0005] In a first aspect, embodiments of the present invention provide a method for analyzing enterprise risk, including:

[0006] Obtain enterprise risk data of the target enterprise, wherein the enterprise risk data includes offline government and enterprise data and specific data, as well as personnel data collected in real time from facial recognition devices;

[0007] The government and enterprise data, the specific data, and the personnel data are input into the enterprise risk early warning model to determine the enterprise risk score of the target enterprise. The enterprise risk score is used to indicate the enterprise risk level of the target enterprise.

[0008] In one possible implementation, the step of inputting the government and enterprise data, the specific data, and the personnel data into the enterprise risk early warning model to determine the enterprise risk score of the target enterprise includes:

[0009] The risk score rules for the target enterprise are determined based on the enterprise risk early warning model.

[0010] Based on the risk score rules, the maximum risk score set for the government and enterprise data, the specific data, and the personnel data under the percentage system is determined, as well as the judgment rules and maximum risk scores for each type of risk.

[0011] The government and enterprise data, the specific data, and the personnel data are analyzed. If the judgment rules for the corresponding risk are met, the corresponding scores are accumulated to obtain the score for the corresponding risk. This score is not greater than the maximum risk score set for the corresponding risk.

[0012] The sum of the integrals of each risk is determined to obtain the risk score of the corresponding data. This risk score is less than the maximum risk score set for the corresponding data.

[0013] The sum of the risk scores of each data point is taken as the enterprise risk score of the target enterprise.

[0014] In one possible implementation, after inputting the government and enterprise data, the specific data, and the personnel data into the enterprise risk early warning model to determine the enterprise risk score of the target enterprise, the method further includes:

[0015] If the enterprise risk score is greater than a preset score threshold, the enterprise risk score will be pushed to the enterprise management organization.

[0016] In one possible implementation, after the enterprise risk score is pushed to the enterprise management organization, the method further includes:

[0017] Receive feedback from the enterprise management structure regarding the modification of the enterprise risk score based on the actual enterprise risk of the target enterprise;

[0018] Based on the feedback results, the risk score rules of the enterprise risk early warning model are adjusted to obtain an adjusted model, so that the risk score rules in the adjusted model conform to the actual enterprise risk.

[0019] Secondly, embodiments of the present invention also provide an enterprise risk analysis device, comprising:

[0020] The acquisition unit is used to acquire enterprise risk data of the target enterprise, wherein the enterprise risk data includes offline government and enterprise data and specific data, as well as personnel data collected in real time from facial recognition devices;

[0021] The determining unit is used to input the government and enterprise data, the specific data, and the personnel data into the enterprise risk early warning model to determine the enterprise risk score of the target enterprise. The enterprise risk score is used to indicate the enterprise risk level of the target enterprise.

[0022] In one possible implementation, the determining unit is used to:

[0023] The risk score rules for the target enterprise are determined based on the enterprise risk early warning model.

[0024] Based on the risk score rules, the maximum risk score set for the government and enterprise data, the specific data, and the personnel data under the percentage system is determined, as well as the judgment rules and maximum risk scores for each type of risk.

[0025] The government and enterprise data, the specific data, and the personnel data are analyzed. If the judgment rules for the corresponding risk are met, the corresponding scores are accumulated to obtain the score for the corresponding risk. This score is not greater than the maximum risk score set for the corresponding risk.

[0026] The sum of the integrals of each risk is determined to obtain the risk score of the corresponding data. This risk score is less than the maximum risk score set for the corresponding data.

[0027] The sum of the risk scores of each data point is taken as the enterprise risk score of the target enterprise.

[0028] In one possible implementation, after the determining unit inputs the government and enterprise data, the specific data, and the personnel data into the enterprise risk early warning model to determine the enterprise risk score of the target enterprise, the analysis device further includes:

[0029] The push unit is used to push the enterprise risk score to the enterprise management organization if the enterprise risk score is greater than a preset score threshold.

[0030] Thirdly, embodiments of the present invention also provide an enterprise risk analysis apparatus, the analysis apparatus including a processor, the processor being configured to execute a computer program stored in a memory to implement the steps of the enterprise risk analysis method as described in any of the above.

[0031] Fourthly, embodiments of the present invention also provide a readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the enterprise risk analysis method as described in any of the preceding claims.

[0032] Fifthly, embodiments of the present invention also provide a computer program product, including computer program instructions that, when executed by a processor, implement the steps of the method as described in any of the above.

[0033] The beneficial effects of this invention are as follows:

[0034] This invention provides a method, apparatus, storage medium, and program product for analyzing enterprise risk. First, it gathers enterprise risk data for the target enterprise, including offline government and enterprise data, specific data, and personnel data collected in real-time from facial recognition devices. Then, it inputs the government and enterprise data, specific data, and personnel data into an enterprise risk early warning model to determine an enterprise risk score indicating the target enterprise's risk level. Because the enterprise risk analysis process considers not only offline government and enterprise data and specific data but also combines real-time personnel data collected by facial recognition devices for comprehensive analysis, the enterprise risk data is more diverse, thereby improving the accuracy of enterprise risk assessment. Furthermore, since the personnel data is collected in real-time by facial recognition devices, it ensures that enterprise management can promptly and effectively detect and issue early warnings of enterprise risks. Attached Figure Description

[0035] Figure 1 A flowchart illustrating a method for analyzing enterprise risk provided in an embodiment of the present invention;

[0036] Figure 2 for Figure 1 Flowchart of one method for step S102;

[0037] Figure 3 A flowchart illustrating one method of analyzing enterprise risk as provided in this embodiment of the invention, after the enterprise risk score is pushed to the enterprise management organization;

[0038] Figure 4 This is a schematic diagram illustrating one aspect of the acquisition and processing of offline government and enterprise data and specific data in an enterprise risk analysis method provided by an embodiment of the present invention.

[0039] Figure 5 This is a schematic diagram illustrating one aspect of the process for acquiring and processing personnel data collected in real time from facial recognition data in an enterprise risk analysis method provided by an embodiment of the present invention.

[0040] Figure 6 This is a schematic diagram of one of the learning and training processes of an enterprise risk early warning model in an enterprise risk analysis method provided by an embodiment of the present invention;

[0041] Figure 7 This is a schematic diagram of one structure of an enterprise risk analysis device provided in an embodiment of the present invention. Detailed Implementation

[0042] In the specification, claims, and accompanying drawings of this invention, the term "comprising" and any variations thereof are intended to cover a non-exclusive inclusion. For example, a process, method, system, product, or apparatus that comprises a series of steps or units is not limited to the steps or units listed, but may optionally include steps or units not listed, or may optionally include other steps or units inherent to such processes, methods, products, or apparatus.

[0043] In this document, the term "embodiment" means that a particular feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of the invention. The appearance of this phrase in various places throughout the specification does not necessarily refer to the same embodiment, nor is it a separate or alternative embodiment mutually exclusive with other embodiments. It will be explicitly and implicitly understood by those skilled in the art that the embodiments described herein can be combined with other embodiments.

[0044] To better understand the above technical solutions, the technical solutions of the present invention will be described in detail below with reference to the accompanying drawings and specific embodiments. It should be understood that the embodiments of the present invention and the specific features in the embodiments are detailed descriptions of the technical solutions of the present invention, rather than limitations on the technical solutions of the present invention. In the absence of conflict, the embodiments of the present invention and the technical features in the embodiments can be combined with each other.

[0045] In related technologies, offline analysis of government and enterprise basic data is often used to analyze enterprise risk characteristics. Specifically, this involves acquiring government and enterprise basic data, including basic business registration information, industry information, legal information, and change information, and using this data as a training set to train a predictive model. After acquiring the government and enterprise basic data of the enterprise to be predicted, this data is input into the predictive model to obtain the risk probability of the enterprise. However, because government and enterprise basic data is relatively limited in variety, and even after data cleaning, the amount of effective enterprise risk data is even smaller, leading to inaccurate model predictions. Furthermore, although machine learning can be used to continuously optimize the predictive model in related technologies, the limited amount of sample data cannot effectively improve the accuracy of enterprise risk prediction.

[0046] In view of this, embodiments of the present invention provide a method, apparatus, storage medium, and program product for analyzing enterprise risks, which can improve the accuracy of enterprise risk assessment.

[0047] like Figure 1 As shown, this embodiment of the invention provides a method for analyzing enterprise risk, including:

[0048] S101: Obtain enterprise risk data of the target enterprise, wherein the enterprise risk data includes offline government and enterprise data and specific data, as well as personnel data collected in real time from facial recognition devices;

[0049] In the specific implementation process, the first step is to acquire the enterprise risk data of the target company. The target company can be an affiliated enterprise of an enterprise management organization. The enterprise management organization can be a department that manages various enterprises within a specific region, such as a financial city or a science park. Alternatively, the enterprise management organization can be an organization defined by its business scope. Of course, the enterprise management organization can also be an organization defined according to actual application needs, without limitation here. Furthermore, the enterprise risk data includes discrete government and enterprise data and specific data, as well as personnel data collected in real time from facial recognition devices. Government and enterprise data includes basic business registration information, industry information, legal information, change information, legal proceedings, and legal information of affiliated enterprises related to the target company. Specific data can be data from specific platforms, including accident personnel, organizations, cases (incidents), locations, and items.

[0050] In one exemplary embodiment, government and enterprise data and specific data may be collected uniformly on a daily schedule. The collected data may also be cleaned to filter out data of enterprises associated with enterprise management agencies, and even to filter out government and enterprise data and specific data related to the target enterprise from these data, and use them as enterprise risk data of the target enterprise.

[0051] In one exemplary embodiment, the personnel data collected by the facial recognition device can also be acquired in real time, and personnel data of relevant personnel associated with the target enterprise can be filtered out. The personnel data can be data obtained by recognizing and analyzing the facial images collected by the facial recognition device through a personnel database, such as gender, age, specific information, etc.

[0052] S102: Input the government and enterprise data, the specific data, and the personnel data into the enterprise risk early warning model to determine the enterprise risk score of the target enterprise. The enterprise risk score is used to indicate the enterprise risk level of the target enterprise.

[0053] After acquiring offline government and enterprise data and specific data, as well as personnel data collected in real time from facial recognition devices, the government and enterprise data, specific data, and personnel data are input into the enterprise risk early warning model to obtain the enterprise risk score of the target enterprise. The enterprise risk score indicates the enterprise risk level of the target enterprise, and the enterprise risk early warning model is a model pre-trained based on a training set. In one exemplary embodiment, the correspondence between the enterprise risk score and the enterprise risk level can be as follows: an enterprise risk score between 90 and 100 points indicates a severely high-risk enterprise; an enterprise risk score between 80 and 100 points indicates a high-risk enterprise; an enterprise risk score between 60 and 80 points indicates a medium-risk enterprise; and an enterprise risk score below 60 points indicates a low-risk enterprise. In this way, the enterprise risk level of the target enterprise can be determined by its determined enterprise risk score, thereby achieving early warning of the enterprise risk of the target enterprise.

[0054] Furthermore, since the enterprise risk analysis method provided in this embodiment of the invention not only considers offline government and enterprise data and specific data, but also needs to combine real-time personnel data collected by facial recognition devices for comprehensive analysis, the enterprise risk data is more diverse, thereby improving the accuracy of enterprise risk assessment. Moreover, because the personnel data is collected in real-time by facial recognition devices, it ensures that enterprise management organizations can promptly and effectively identify enterprise risks and issue timely warnings.

[0055] In one exemplary embodiment, such as Figure 2 As shown, step S102: Inputting the government and enterprise data, the specific data, and the personnel data into the enterprise risk early warning model to determine the enterprise risk score of the target enterprise, including:

[0056] S201: Determine the risk score rules for the target enterprise based on the enterprise risk early warning model;

[0057] S202: Based on the risk score rules, determine the maximum risk score set for the government and enterprise data, the specific data, and the personnel data respectively under the percentage system, as well as the judgment rules and maximum risk scores for each risk set respectively;

[0058] S203: Analyze the government and enterprise data, the specific data, and the personnel data. If the judgment rules for the corresponding risk are met, accumulate the corresponding points to obtain the points for the corresponding risk. The points shall not be greater than the maximum risk points set for the corresponding risk.

[0059] S204: Determine the sum of the integrals of each risk to obtain the risk score of the corresponding data. This risk score is less than the maximum risk score set for the corresponding data.

[0060] S205: The sum of the risk scores of each data point shall be taken as the enterprise risk score of the target enterprise.

[0061] In the specific implementation process, steps S201 to S205 are implemented as follows:

[0062] First, the risk scoring rules for the target enterprise are determined based on the enterprise risk warning model. Then, based on the risk scoring rules, the maximum risk score is determined for government and enterprise data, specific data, and personnel data under a 100-point scale, along with the judgment rules and maximum risk score for each risk category. For example, using a uniform 100-point scale, the maximum risk score for government and enterprise data is 50 points, for specific data it is 20 points, and for personnel data it is 30 points. Personnel data includes two risks: predatory lending and loans to the elderly. The maximum risk score for both predatory lending and loans to the elderly is 15 points. The judgment rule for loans to the elderly is that the enterprise is labeled as internet finance and the number of people over 60 years old who have entered the platform in the past three days exceeds the abnormal value. The judgment rule for predatory lending is that the person who reported the incident has a record of reporting fraud.

[0063] Then, the government and enterprise data, specific data, and personnel data are analyzed. If the judgment rules for the corresponding risk are met, the corresponding points are accumulated to obtain the corresponding risk score. This score cannot exceed the maximum risk score set for the corresponding risk. Taking the aforementioned elderly loan as an example, the facial recognition device at the entrance of Floor A in the Financial City Science and Technology Park has accumulated dozens of people over the age of 60 in the past three days, far exceeding the proportion of people in the same age group in previous periods. Combined with the offline analysis showing that the enterprise risk label is "Internet Metal," it is judged that the abnormal entry and exit of elderly people in the past three days may indicate an elderly loan issue. The offline data is then added to determine if there are any financial-related alarm records among the elderly people entering and exiting, and whether there are any records of financial fraud or complaints. This accumulation of points calculates the corresponding risk score for the elderly loan. The accumulated score for this risk cannot exceed the set 15 points. In other words, the maximum accumulated score for the elderly loan risk is 15 points. Using the same point accumulation calculation method, by analyzing government and enterprise data, specific data, and personnel data, the scores for each risk corresponding to each data point can be determined.

[0064] After determining the scores for each risk, the sum of these scores can be calculated to obtain the risk score for the corresponding data. This risk score is less than the maximum risk score set for that data. Using the aforementioned personnel data as an example, if the risk score for predatory lending is 10 and the risk score for senior citizen loans is 5 after cumulative calculation, then the risk score for the personnel data is 15 points. Using a similar calculation method as for personnel data, the risk scores for government and enterprise data, as well as specific data, can be determined. Then, the sum of the risk scores for all data is used as the target company's enterprise risk score. For example, if the target company's government and enterprise data risk score is 40 points, the specific data risk score is 10 points, and the personnel data risk score is 15 points, then the target company's enterprise risk score is 65 points. Based on the correspondence between enterprise risk scores and enterprise risk levels, the target company's risk level can be determined to be medium risk.

[0065] In this embodiment of the invention, after step S102: inputting the government and enterprise data, the specific data, and the personnel data into the enterprise risk early warning model to determine the enterprise risk score of the target enterprise, the method further includes:

[0066] If the enterprise risk score is greater than a preset score threshold, the enterprise risk score will be pushed to the enterprise management organization.

[0067] In practice, after inputting government and enterprise data, specific data, and personnel data into the enterprise risk early warning model to determine the target enterprise's risk score, if the enterprise risk score exceeds a preset threshold, the score can be pushed to the enterprise management organization, thereby enabling timely early warning of the target enterprise's risks by the management structure. Furthermore, relevant personnel in the enterprise management organization can view detailed information about the target enterprise's risks; for example, in addition to displaying the target enterprise's risk score, they can also see the risk type and the corresponding risk score for each risk.

[0068] In embodiments of the present invention, such as Figure 3 As shown, after step 1: pushing the enterprise risk score to the enterprise management organization, the method further includes:

[0069] S301: Receive feedback from the enterprise management structure that it has modified the enterprise risk score based on the actual enterprise risk of the target enterprise;

[0070] S302: Adjust the risk score rules of the enterprise risk early warning model according to the feedback results to obtain the adjusted model, so that the risk score rules in the adjusted model conform to the actual enterprise risk.

[0071] In the specific implementation process, steps S301 to S302 are implemented as follows:

[0072] First, the system receives feedback from the enterprise management structure regarding modifications to the enterprise risk score based on the target enterprise's actual risks. For example, in specific data, although there may be complaints against individuals associated with the target enterprise, these complaints may be clearly unreasonable upon verification. However, due to flawed risk assessment rules in the enterprise risk warning model, the obtained enterprise risk score may be significantly too high. After investigation by the enterprise management, the predicted enterprise risk score can be modified based on the target enterprise's actual risks, thus receiving the corresponding feedback. Following this feedback, the risk score rules of the enterprise risk warning model can be adjusted, resulting in an adjusted enterprise risk warning model. Because the enterprise management can modify the predicted enterprise risk score based on new risks, the enterprise risk warning model can learn the latest risk score rules, thereby adjusting the maximum risk score for each data point in risk prediction, as well as the judgment rules and maximum risk scores for each risk corresponding to each data point. In this way, the enterprise risk warning model is dynamically adjusted through feedback from the enterprise management, ensuring its accuracy and improving the precision of subsequent enterprise risk predictions.

[0073] It should be noted that Natural Language Processing (NLP) can be used to adjust the enterprise risk warning model based on the feedback results. This allows for optimization of the model, especially since many enterprise risk characteristics are highly specialized, through confirmation by relevant professionals within the enterprise's management structure. Furthermore, because different enterprise management organizations may have varying levels of focus on different risk types, modifications by these organizations ensure differentiated assessments of the warning model, thereby improving the practicality of enterprise risk prediction.

[0074] It should be noted that the enterprise risk analysis method provided in this embodiment of the invention can be applied to an enterprise risk early warning platform that facilitates enterprise management agencies in monitoring enterprises. The acquisition and processing of offline government and enterprise data and specific data can be as follows: Figure 4 As shown. Specifically, firstly, offline government and enterprise data and specific data can be aggregated and collected using data aggregation tools. Some data can be obtained from relational databases such as MySQL, while some data can be collected from external systems through scheduled tasks (Jobs). For example, if the external system is an alarm platform, the corresponding external system data can be obtained from the alarm platform. Of course, external systems can be selected and the collected external system data can be filtered according to actual application needs; no restrictions are imposed here.

[0075] After data collection, the relevant data can be cleaned according to preset data cleaning rules, such as Kettle, to filter out government and enterprise data and specific data related to the target enterprise, while filtering out data unrelated to enterprise risk, thus ensuring efficient data storage. Then, the cleaned data can be stored in a big data analytics database such as ClickHouse using a distributed message queue, where the distributed message queue could be Pulsar. Next, offline analysis technology using Spark, combined with big data processing capabilities such as Yarn, can be used to calculate risk scores based on an enterprise risk warning model. This, combined with enterprise tagging, facilitates the rapid definition and analysis of enterprise risks. For example, enterprise tags could include internet finance, predatory lending, frequent complaints, and high-risk related enterprises.

[0076] The process of acquiring and processing personnel data collected in real time from facial recognition devices can be as follows: Figure 5 As shown. Specifically, firstly, facial recognition devices can be connected to the enterprise risk warning platform via various protocols such as MQTT and HTTP / 1. This allows the facial recognition devices to report real-time captured facial images to the platform. Upon receiving the report, the platform can parse the message and push it to a distributed message queue, such as the Pulsar distributed message queue. After retrieving the message from the distributed message queue, the platform can query the corresponding personnel data for the facial image, such as gender, age, and specific records, using a pre-defined personnel database.

[0077] In practical implementation, the system can combine real-time analysis of age distribution and number of people with offline analysis of enterprise risk tags to comprehensively determine whether the enterprise meets the corresponding risk rules in the enterprise risk warning model. If it does, the risk score of the enterprise associated with the facial recognition device is accumulated. In addition, enterprises can be classified into risk-based categories. For example, in the Financial City Science and Technology Park, the facial recognition device at the entrance of a certain floor has accumulated dozens of people over the age of 60 in the past three days, far exceeding the population proportion of previous time periods. Combined with the enterprise risk tags such as Internet finance in the offline analysis, it can be determined that the abnormal entry and exit of elderly people in the past three days may indicate the presence of senior citizen loans. In addition, it can accumulate specific offline data of elderly people entering and exiting the area to see if there are any financial-related alarm records, or whether there are any records of financial fraud or complaints. Thus, by combining real-time and offline data, the possibility of new types of enterprise risks such as senior citizen loans can be comprehensively assessed.

[0078] Furthermore, the learning and training process of an enterprise risk warning model can be as follows: Figure 6As shown. First, the risk score rules of the initial enterprise risk warning model can be debugged and adjusted using existing data to set the proportions of risk modules, namely the proportions of government and enterprise data, specific data, and personnel data in enterprise risk assessment. Individual risk scores can also be set according to risk types. Subsequently, feedback from enterprise management regarding the actual enterprise risk situation and modifications can be used for machine learning training to learn the latest risk score rules. This continuously adjusts and optimizes the proportions of risk modules, improves individual warning rules and individual risk scores, and then feeds back the latest risk warning model, resulting in an optimized enterprise risk warning model that ensures the accuracy of enterprise risk warnings.

[0079] In one exemplary embodiment, a comprehensive risk score can be calculated based on the risk type, volume, and degree of the accessed data. This score is then accumulated according to one-way risk rules, with the total score not exceeding the maximum score of the corresponding proportion. Later, the data sources for enterprise risk can be dynamically increased, and machine learning can be used to continuously adjust the module proportions and individual rule risk scores. For example, if specific data includes illegal activities by key personnel of an enterprise, and this meets the enterprise risk warning model, risk scores are accumulated for the related enterprise. Similarly, if a thorough investigation of key personnel in government and enterprise data reveals numerous negative reports about related enterprises, and this meets the enterprise risk warning model, risk scores are accumulated for the related enterprises. Of course, risk scores can also be accumulated for specific enterprises according to actual application needs, which will not be detailed here.

[0080] Based on the same inventive concept, such as Figure 7 As shown, this embodiment of the invention also provides a device for analyzing enterprise risk, including:

[0081] The acquisition unit 10 is used to acquire enterprise risk data of the target enterprise, wherein the enterprise risk data includes offline government and enterprise data and specific data, as well as personnel data collected in real time from the face recognition device;

[0082] The determining unit 20 is used to input the government and enterprise data, the specific data and the personnel data into the enterprise risk early warning model to determine the enterprise risk score of the target enterprise. The enterprise risk score is used to indicate the enterprise risk level of the target enterprise.

[0083] In this embodiment of the invention, the determining unit 20 is used for:

[0084] The risk score rules for the target enterprise are determined based on the enterprise risk early warning model.

[0085] Based on the risk score rules, the maximum risk score set for the government and enterprise data, the specific data, and the personnel data under the percentage system is determined, as well as the judgment rules and maximum risk scores for each type of risk.

[0086] The government and enterprise data, the specific data, and the personnel data are analyzed. If the judgment rules for the corresponding risk are met, the corresponding scores are accumulated to obtain the score for the corresponding risk. This score is not greater than the maximum risk score set for the corresponding risk.

[0087] The sum of the integrals of each risk is determined to obtain the risk score of the corresponding data. This risk score is less than the maximum risk score set for the corresponding data.

[0088] The sum of the risk scores of each data point is taken as the enterprise risk score of the target enterprise.

[0089] In this embodiment of the invention, after the determining unit inputs the government and enterprise data, the specific data, and the personnel data into the enterprise risk early warning model to determine the enterprise risk score of the target enterprise, the analysis device further includes:

[0090] The push unit is used to push the enterprise risk score to the enterprise management organization if the enterprise risk score is greater than a preset score threshold.

[0091] Based on the same inventive concept, embodiments of the present invention also provide an enterprise risk analysis apparatus, the analysis apparatus including a processor, the processor being configured to execute a computer program stored in a memory to implement the steps of the enterprise risk analysis method as described in any of the above claims.

[0092] Based on the same inventive concept, embodiments of the present invention also provide a readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the enterprise risk analysis method as described in any of the above claims.

[0093] Based on the same inventive concept, embodiments of the present invention also provide a computer program product, including computer program instructions that, when executed by a processor, implement the steps of any of the methods described above.

[0094] This invention provides a method, apparatus, storage medium, and program product for analyzing enterprise risk. First, it gathers enterprise risk data for the target enterprise, including offline government and enterprise data, specific data, and personnel data collected in real-time from facial recognition devices. Then, it inputs the government and enterprise data, specific data, and personnel data into an enterprise risk early warning model to determine an enterprise risk score indicating the target enterprise's risk level. Because the enterprise risk analysis process considers not only offline government and enterprise data and specific data but also combines real-time personnel data collected by facial recognition devices for comprehensive analysis, the enterprise risk data is more diverse, thereby improving the accuracy of enterprise risk assessment. Furthermore, since the personnel data is collected in real-time by facial recognition devices, it ensures that enterprise management can promptly and effectively detect and issue early warnings of enterprise risks.

[0095] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0096] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to this application. It should be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0097] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0098] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0099] Obviously, those skilled in the art can make various modifications and variations to this application without departing from the spirit and scope of this application. Therefore, if such modifications and variations fall within the scope of the claims of this application and their equivalents, this application also intends to include such modifications and variations.

Claims

1. A method for analyzing enterprise risk, characterized in that, include: The system acquires enterprise risk data for a target company, including offline government and enterprise data, specific data, and personnel data collected in real time from facial recognition devices. The target company is an affiliated enterprise of a corporate management organization. The facial recognition devices are deployed within the scope of the corporate management organization and correspond to the target company's business premises. The government and enterprise data includes basic business registration information, industry information, legal information, change information, legal proceedings, and legal information of affiliated companies related to the target company. The specific data comes from a specific platform and includes information on personnel, organizations, events, locations, and items involved in the incident. The personnel data includes gender, age, and specific information. The government and enterprise data, the specific data, and the personnel data are input into the enterprise risk early warning model to determine the enterprise risk score of the target enterprise. The enterprise risk score is used to indicate the enterprise risk level of the target enterprise. The enterprise risk early warning model is a model used to predict new types of enterprise risks, including loans to the elderly and predatory lending. The step of inputting the government and enterprise data, the specific data, and the personnel data into the enterprise risk early warning model to determine the enterprise risk score of the target enterprise includes: The risk score rules for the target enterprise are determined based on the enterprise risk early warning model. Based on the risk score rules, the maximum risk score set for the government and enterprise data, the specific data, and the personnel data under the percentage system is determined, as well as the judgment rules and maximum risk scores for each type of risk. The government and enterprise data, the specific data, and the personnel data are analyzed. If the judgment rules for the corresponding risk are met, the corresponding points are accumulated to obtain the score for the corresponding risk. This score is not greater than the maximum risk score set for the corresponding risk. The sum of the scores for each risk is determined to obtain the risk score for the corresponding data. This risk score is less than the maximum risk score set for the corresponding data. The sum of the risk scores of each data point is taken as the enterprise risk score of the target enterprise. The data sources for dynamically increasing enterprise risk are adjusted by using machine learning to continuously adjust the proportions of various items and individual rule risk scores in the government and enterprise data, the specific data, and the personnel data. Receive feedback from the enterprise management structure regarding the modification of the enterprise risk score based on the actual enterprise risk of the target enterprise; Based on the feedback results, the risk score rules of the enterprise risk early warning model are adjusted to obtain an adjusted model, so that the risk score rules in the adjusted model conform to the actual enterprise risk.

2. The method as described in claim 1, characterized in that, After inputting the government and enterprise data, the specific data, and the personnel data into the enterprise risk early warning model to determine the enterprise risk score of the target enterprise, the method further includes: If the enterprise risk score is greater than a preset score threshold, the enterprise risk score will be pushed to the enterprise management organization.

3. A device for analyzing enterprise risk, characterized in that, include: The acquisition unit is used to acquire enterprise risk data of a target enterprise. This enterprise risk data includes offline government and enterprise data and specific data, as well as personnel data collected in real-time from facial recognition devices. The target enterprise is an affiliated enterprise of an enterprise management organization. The facial recognition devices are deployed within the scope of the enterprise management organization and correspond to the target enterprise's business premises. The government and enterprise data includes basic business registration information, industry information, legal information, change information, legal proceedings, and legal information of affiliated enterprises related to the target enterprise. The specific data is data from a specific platform, including accident personnel, organizations, events, locations, and items. The personnel data includes gender, age, and specific information. The determining unit is used to input the government and enterprise data, the specific data, and the personnel data into the enterprise risk early warning model to determine the enterprise risk score of the target enterprise. The enterprise risk score is used to indicate the enterprise risk level of the target enterprise. The enterprise risk early warning model is a model used to predict new types of enterprise risks, including loans to the elderly and predatory lending. The determining unit is used for: The risk score rules for the target enterprise are determined based on the enterprise risk early warning model. Based on the risk score rules, the maximum risk score set for the government and enterprise data, the specific data, and the personnel data under the percentage system is determined, as well as the judgment rules and maximum risk scores for each type of risk. The government and enterprise data, the specific data, and the personnel data are analyzed. If the judgment rules for the corresponding risk are met, the corresponding scores are accumulated to obtain the score for the corresponding risk. This score is not greater than the maximum risk score set for the corresponding risk. The sum of the integrals of each risk is determined to obtain the risk score of the corresponding data. This risk score is less than the maximum risk score set for the corresponding data. The sum of the risk scores of each data point is taken as the enterprise risk score of the target enterprise. The determining unit is further configured to: The data sources for dynamically increasing enterprise risk are adjusted by using machine learning to continuously adjust the proportions of various items and individual rule risk scores in the government and enterprise data, the specific data, and the personnel data. The analytical device is also used for: Receive feedback from the enterprise management structure regarding the modification of the enterprise risk score based on the actual enterprise risk of the target enterprise; Based on the feedback results, the risk score rules of the enterprise risk early warning model are adjusted to obtain an adjusted model, so that the risk score rules in the adjusted model conform to the actual enterprise risk.

4. The analytical apparatus as described in claim 3, characterized in that, After the determining unit inputs the government and enterprise data, the specific data, and the personnel data into the enterprise risk early warning model to determine the enterprise risk score of the target enterprise, the analysis device further includes: The push unit is used to push the enterprise risk score to the enterprise management organization if the enterprise risk score is greater than a preset score threshold.

5. A device for analyzing enterprise risk, characterized in that, The analysis apparatus includes a processor that executes a computer program stored in a memory to implement the steps of the enterprise risk analysis method as described in claim 1 or 2.

6. A readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the enterprise risk analysis method as described in claim 1 or 2.

7. A computer program product comprising computer program instructions, characterized in that, When the computer program instructions are executed by the processor, they implement the steps of the method described in claim 1 or 2.

Images