Risk identification management system and method for investment data

Abstract

The invention provides a risk identification management system and method for investment data, and the system comprises an investment data obtaining subsystem which is used for obtaining the investment data of a management enterprise; the enterprise portrait acquisition subsystem is used for acquiring an enterprise portrait of an investment enterprise; the risk identification scheme library construction subsystem is used for acquiring a risk identification scheme library; the risk identification template determination subsystem is used for determining a risk identification template according to the enterprise portrait and a risk identification scheme library; the risk identification result acquisition subsystem is used for acquiring a risk identification result by comparing the risk identification template based on the investment data; and the risk management subsystem is used for carrying out risk management on the investment enterprise according to the risk identification result. According to the risk identification management system and method for investment data, a risk identification template is formulated according to an enterprise portrait and a risk identification scheme library; the investment data is compared with the risk identification template, the risk identification result is obtained and managed, and the risk monitoring is more standard and timely.

Description

Technical Field

[0001] The present invention relates to the technical field of enterprise management, and in particular to a risk identification management system and method for investment data. Background Art

[0002] Risk management is an important part of business operations. It helps companies comprehensively identify, evaluate and manage various potential risks. Through advanced data analysis and real-time monitoring functions, it can promptly discover risk factors that may affect business operations, conduct in-depth analysis, and evaluate their likelihood and potential impact.

[0003] The invention patent with application number: CN201810236541.2 discloses an investment audit management and analysis system, including a financial acquisition module, a risk assessment module, an investment assessment module and an investment diversion module; the financial acquisition module is used to obtain the operating information of the invested enterprise and the current income and expenditure status obtained by the investment, and send it to the risk assessment module in the order of the total income and expenditure difference; the risk assessment module receives the information sent by the financial acquisition module to determine the risk coefficient and return level coefficient of the investment, and sends it to the investment assessment module; the investment assessment module is used to judge the comprehensive investment coefficient of each invested enterprise, and output the comprehensive investment coefficient to the investment diversion module. The above invention calculates the risk coefficient and return level coefficient of the invested enterprise through the risk assessment module, and evaluates the comprehensive investment coefficient through the investment assessment module to screen the invested enterprises with higher comprehensive investment coefficients, which helps managers to select invested enterprises and improves the efficiency of investment audits.

[0004] However, the corporate structures of the invested enterprises are different, and the appropriate assessment schemes for different enterprises are also different. If the unified risk assessment module of the above invention is used for assessment, the suitability of risk assessment is not high, and further, the timeliness of risk monitoring and early warning is delayed.

[0005] In view of this, there is an urgent need for a risk identification management system and method for investment data to at least solve the above-mentioned deficiencies. Summary of the invention

[0006] One of the purposes of the present invention is to provide a risk identification management system and method for investment data, which obtains standardized corporate portraits of investment enterprises based on the analysis of investment data. At the same time, a risk identification solution library is introduced, and a risk identification template corresponding to the target risk identification solution is formulated based on the matching results of the corporate portraits and the risk identification solution library; the investment data is compared with the risk identification template, the risk identification results are obtained and managed, thereby improving the standardization and timeliness of risk monitoring of investment enterprises.

[0007] An embodiment of the present invention provides a risk identification and management system for investment data, comprising:

[0008] The investment data acquisition subsystem is used to acquire the investment data of the management enterprise;

[0009] The enterprise portrait acquisition subsystem is used to analyze investment data and obtain the enterprise portrait of the invested enterprise;

[0010] The risk identification solution library construction subsystem is used to obtain the risk identification solution library;

[0011] The risk identification template determination subsystem is used to determine the risk identification template corresponding to the investment enterprise based on the enterprise profile and risk identification solution library;

[0012] The risk identification result acquisition subsystem is used to obtain risk identification results based on investment data and risk identification templates;

[0013] The risk management subsystem is used to conduct risk management on investment enterprises based on risk identification results.

[0014] Preferably, the enterprise portrait acquisition subsystem includes:

[0015] An enterprise portrait description template acquisition module is used to obtain an enterprise portrait description template;

[0016] The enterprise portrait acquisition module is used to obtain the enterprise portrait of the invested enterprise based on the enterprise portrait description template and investment data.

[0017] Preferably, the risk identification solution library construction subsystem includes:

[0018] Enterprise risk event acquisition module, used to acquire enterprise risk events based on target big data nodes;

[0019] An event analysis module is used to analyze enterprise risk events and obtain enterprise descriptions of risky enterprises and risk event-related description items;

[0020] The risk identification scheme determination module is used to obtain the identification sub-strategies of the risk event associated description items, summarize the identification sub-strategies corresponding to the risk event associated description items corresponding to the same risk event, and obtain the risk identification scheme corresponding to the enterprise description;

[0021] The module for determining items to be entered is used to associate the enterprise description with the risk identification plan to obtain items to be entered;

[0022] The construction module is used to enter all items to be entered into the database and obtain a risk identification solution library.

[0023] Preferably, the event analysis module includes:

[0024] The first item to be classified acquisition submodule is used to analyze the enterprise risk event and obtain the first item to be classified; the first item to be classified is: a one-to-one corresponding description of the enterprise to be classified and a description item associated with the first risk event to be classified;

[0025] The second item to be classified acquisition submodule is used to acquire the second item to be classified according to the first item to be classified based on the first classification rule corresponding to the enterprise description; the second item to be classified is: a one-to-one corresponding enterprise description and a description item associated with the second risk event to be classified;

[0026] The classification result acquisition submodule is used to obtain the classification result based on the second classification rule corresponding to the enterprise risk event and the second item to be classified; the classification result includes: one-to-one corresponding enterprise description, risk event identifier and risk event related description item corresponding to the risk event identifier.

[0027] Preferably, the first to-be-classified item acquisition submodule includes:

[0028] A pre-selected first item to be classified obtaining unit is used to analyze the enterprise risk event and obtain the pre-selected first item to be classified;

[0029] An extraction source acquisition unit, used to parse the first item to be classified and acquire the extraction source of the pre-selected risk event related description item;

[0030] A credibility determination unit, used to determine the credibility of the pre-selected risk event associated description items according to the extraction source;

[0031] The first item to be classified determining unit is used to use the pre-selected risk event associated description item of the corresponding extraction source as the risk event associated description item if the credibility is greater than or equal to a preset credibility threshold, and determine the first item to be classified according to the risk event associated description item.

[0032] Preferably, the credibility determination unit includes:

[0033] A first event attribute acquisition subunit, used to acquire a first event attribute of an enterprise risk event;

[0034] A second event attribute acquisition subunit, used to acquire a second event attribute of an extraction source;

[0035] A first credibility determination value determination subunit, configured to calculate the event attribute matching degree according to the first event attribute and the second event attribute, and use the matching degree as the first credibility determination value;

[0036] A correlation information calculation subunit, used to calculate the correlation degree and correlation direction according to the extraction source and the event party of the enterprise risk event;

[0037] An adjustment factor determination subunit, used to determine an adjustment factor of the first credibility determination value according to the correlation degree and the correlation direction;

[0038] A second credibility determination value determination subunit, configured to determine a second credibility determination value according to the adjustment factor and the first credibility determination value;

[0039] A credibility check value acquisition subunit is used to acquire a credibility check value of an extraction source;

[0040] A correction coefficient determination subunit, used to determine a correction coefficient of a second credibility determination value according to the credibility check value and the average credibility check value;

[0041] The credibility acquisition subunit is used to correct the second credibility determination value according to the correction coefficient to obtain the credibility of the extraction source corresponding to the pre-selected risk event related description item.

[0042] Preferably, the risk identification result acquisition subsystem includes:

[0043] The risk data acquisition module is used to obtain the risk data of the investment enterprise according to the identification basis of the risk identification template;

[0044] The risk identification result acquisition module is used to obtain risk identification results based on risk data and risk identification templates.

[0045] Preferably, the risk management subsystem includes:

[0046] The investment behavior dynamic determination module is used to dynamically determine the investment behavior based on the risk identification results.

[0047] Preferably, the investment behavior dynamic determination module includes:

[0048] The rectification plan acquisition submodule is used to obtain the rectification plan of the investment enterprise based on the risk identification results;

[0049] The investment behavior determination submodule is used to monitor the rectification results of the rectification plan and determine the investment behavior.

[0050] Preferably, the rectification plan acquisition submodule includes:

[0051] A historical rectification enterprise determination unit is used to determine historical rectification enterprises whose rectification effects are greater than or equal to a preset rectification effect threshold according to the risk identification result;

[0052] A historical rectification plan acquisition unit, used to acquire historical rectification plans of historical rectification enterprises corresponding to risk identification results;

[0053] A first enterprise characteristic matrix acquisition unit, used to acquire a first enterprise characteristic matrix of a historical rectification enterprise;

[0054] A first local correlation matrix determining unit, used to determine a first local correlation matrix in a first enterprise characteristic matrix according to a historical rectification plan;

[0055] A second enterprise characteristic matrix determining unit, used to determine a second enterprise characteristic matrix of the venture capital enterprise;

[0056] A second local correlation matrix determining unit, used to determine a second local correlation matrix in the second enterprise characteristic matrix according to the relative positions of the first local correlation matrix and the first enterprise characteristic matrix;

[0057] A first matrix similarity calculation unit, used for calculating the first matrix similarity of the first local correlation matrix and the second local correlation matrix;

[0058] A second matrix similarity calculation unit is used to calculate a second matrix similarity of a first local matrix excluding the first local correlation matrix of the first enterprise feature matrix and a second local matrix excluding the second local correlation matrix of the second enterprise feature matrix if the first matrix similarity is greater than or equal to a preset first matrix similarity threshold;

[0059] A training rectification plan determination unit, configured to use the corresponding historical rectification plan as the training rectification plan if the second matrix similarity is greater than or equal to a preset second matrix similarity threshold; wherein the first matrix similarity threshold is greater than the second matrix similarity threshold;

[0060] The rectification plan acquisition unit is used to train a risk rectification model corresponding to the risk identification result according to the training rectification plan, input the second enterprise feature matrix into the risk rectification model, and obtain the rectification plan.

[0061] An embodiment of the present invention provides a risk identification and management method for investment data, comprising:

[0062] Obtain investment data of management enterprises;

[0063] Analyze investment data and obtain corporate portraits of invested companies;

[0064] Obtain a library of risk identification solutions;

[0065] Determine the risk identification template corresponding to the investment enterprise based on the enterprise portrait and risk identification solution library;

[0066] Obtain risk identification results based on investment data and risk identification templates;

[0067] Conduct risk management for investment enterprises based on risk identification results.

[0068] Preferably, the investment data is analyzed to obtain the corporate profile of the invested enterprise, including:

[0069] Get the enterprise portrait description template;

[0070] Obtain the corporate profile of the invested enterprise based on the corporate profile description template and investment data.

[0071] The beneficial effects of the present invention are:

[0072] The present invention obtains standardized corporate portraits of investment enterprises based on parsed investment data. At the same time, a risk identification solution library is introduced. According to the matching results of the corporate portraits and the risk identification solution library, a risk identification template corresponding to the target risk identification solution is formulated; the investment data is compared with the risk identification template, the risk identification results are obtained and managed, thereby improving the standardization and timeliness of risk monitoring of investment enterprises.

[0073] Other features and advantages of the present invention will be described in the following description, and partly become apparent from the description, or understood by practicing the present invention. The purpose and other advantages of the present invention can be realized and obtained by the structures specifically pointed out in the present application documents.

[0074] The technical solution of the present invention is further described in detail below through the accompanying drawings and embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0075] The accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation of the present invention. In the accompanying drawings:

[0076] Figure 1 A schematic diagram of a risk identification and management system for investment data in an embodiment of the present invention;

[0077] Figure 2 It is a schematic diagram of a risk identification and management method for investment data in an embodiment of the present invention. DETAILED DESCRIPTION

[0078] The preferred embodiments of the present invention are described below in conjunction with the accompanying drawings. It should be understood that the preferred embodiments described herein are only used to illustrate and explain the present invention, and are not used to limit the present invention.

[0079] The embodiment of the present invention provides a risk identification and management system for investment data, such as Figure 1 As shown, including:

[0080] The investment data acquisition subsystem 1 is used to acquire the investment data of the management enterprise;

[0081] Enterprise portrait acquisition subsystem 2, used to analyze investment data and obtain enterprise portraits of investment enterprises;

[0082] The risk identification solution library construction subsystem 3 is used to obtain the risk identification solution library;

[0083] The risk identification template determination subsystem 4 is used to determine the risk identification template corresponding to the investment enterprise according to the enterprise profile and the risk identification solution library;

[0084] The risk identification result acquisition subsystem 5 is used to obtain the risk identification result based on the investment data and the risk identification template;

[0085] The risk management subsystem 6 is used to carry out risk management on the investment enterprise according to the risk identification results.

[0086] The working principle and beneficial effects of the above technical solution are:

[0087] The management enterprise has a network communication node that interacts with the outside world, namely, a data docking node. The above data docking node is docked to obtain investment data. The investment data includes the enterprise profile (such as name, industry, registered capital), enterprise qualifications (such as financial operations, scientific research technology, production technology and quality management), enterprise performance record (such as enterprise contract execution and performance), enterprise risk credit (such as judicial risk, operating risk), enterprise social behavior responsibility (such as environmental protection behavior and responsibility, low-carbon efficiency) and enterprise ecological innovation of the investment enterprise of the management enterprise. Obtain the enterprise portrait description template; according to the enterprise portrait description template and investment data, obtain the enterprise portrait of the investment enterprise. The enterprise portrait description template is a template that constrains only the extraction of relevant information about the enterprise description, and the enterprise portrait is the description result of the enterprise; the risk identification solution library includes: one-to-one corresponding enterprise descriptions and corresponding risk identification solutions for the described enterprises. Match the enterprise portrait with the enterprise description in the risk identification solution library, obtain the target risk identification solution corresponding to the matching enterprise description, and formulate the corresponding risk identification template. The risk identification template is a framework formulated based on the target risk identification solution to guide the risk identification process. Based on the identification basis of the risk identification template, the risk data of the investment enterprise is obtained; based on the risk data, the risk identification template is compared to obtain the risk identification results (for example, if the investment enterprise has recent operating risks, a risk reminder is issued). Managers can determine whether to continue investing in the same enterprise based on the risk identification results and adjust the investment strategy.

[0088] The present invention obtains standardized corporate portraits of investment enterprises based on parsed investment data. At the same time, a risk identification solution library is introduced. According to the matching results of the corporate portraits and the risk identification solution library, a risk identification template corresponding to the target risk identification solution is formulated; the investment data is compared with the risk identification template, the risk identification results are obtained and managed, thereby improving the standardization and timeliness of risk monitoring of investment enterprises.

[0089] In one embodiment, the risk identification solution library construction subsystem includes:

[0090] Enterprise risk event acquisition module, used to acquire enterprise risk events based on target big data nodes;

[0091] An event analysis module is used to analyze enterprise risk events and obtain enterprise descriptions of risky enterprises and risk event-related description items;

[0092] The risk identification scheme determination module is used to obtain the identification sub-strategies of the risk event associated description items, summarize the identification sub-strategies corresponding to the risk event associated description items corresponding to the same risk event, and obtain the risk identification scheme corresponding to the enterprise description;

[0093] The module for determining items to be entered is used to associate the enterprise description with the risk identification plan to obtain items to be entered;

[0094] The construction module is used to enter all items to be entered into the database and obtain a risk identification solution library.

[0095] The working principle and beneficial effects of the above technical solution are:

[0096] When constructing a risk identification solution library, it can be constructed based on the historical data of risk identification and warning events that have occurred. When constructing, the target big data node is introduced. The target big data node is used to capture the enterprise risk events that have occurred. Based on different enterprise characteristics, the corresponding enterprise risk events are also different. Therefore, the enterprise risk events are analyzed to obtain the enterprise description of the risk enterprise and the risk event related description items. The risk event related description items are descriptive information related to the risk event, such as: causes, event types and impacts. The risk event related description items contain key information for risk event identification, such as: the risk event related description item is: "Company A is insolvent. It would be great if we monitored its financial statements in time and found data anomalies." The identification sub-strategy is: monitoring financial statements; the risk identification solution is: a summary of risk identification strategies corresponding to different risk events for different enterprise descriptions.

[0097] The present invention introduces a target big data node, obtains enterprise risk events, and determines the enterprise description of the risk enterprise and the risk event associated description items, and determines the identification sub-strategy of the risk enterprise corresponding to the corresponding risk event according to the risk event associated description; summarizes the identification sub-strategies corresponding to the risk event associated description items of the same risk event corresponding to the same enterprise description, obtains the risk identification scheme for each enterprise description corresponding to different risk events under the enterprise description, associates the enterprise description with the risk identification scheme, obtains the items to be entered and puts them into the database, and obtains the risk identification scheme library, thereby improving the comprehensiveness and rationality of the construction of the risk identification scheme library.

[0098] In one embodiment, the event parsing module includes:

[0099] The first item to be classified acquisition submodule is used to analyze the enterprise risk event and obtain the first item to be classified; the first item to be classified is: a one-to-one corresponding description of the enterprise to be classified and a description item associated with the first risk event to be classified;

[0100] The second item to be classified acquisition submodule is used to acquire the second item to be classified according to the first item to be classified based on the first classification rule corresponding to the enterprise description; the second item to be classified is: a one-to-one corresponding enterprise description and a description item associated with the second risk event to be classified;

[0101] The classification result acquisition submodule is used to obtain the classification result based on the second classification rule corresponding to the enterprise risk event and the second item to be classified; the classification result includes: one-to-one corresponding enterprise description, risk event identifier and risk event related description item corresponding to the risk event identifier.

[0102] The working principle and beneficial effects of the above technical solution are:

[0103] The present invention introduces a first classification rule and a second classification rule, performs a primary classification of risk event-related description items according to the enterprise description, and in each enterprise description, performs a secondary classification of risk event-related description items according to the enterprise risk events; the classified analysis results are used for the formulation of subsequent risk identification plans, further improving the accuracy of the formulation of risk identification plans.

[0104] In one embodiment, the first to-be-classified item acquisition submodule includes:

[0105] A pre-selected first item to be classified obtaining unit is used to analyze the enterprise risk event and obtain the pre-selected first item to be classified;

[0106] An extraction source acquisition unit, used to parse the first item to be classified and acquire the extraction source of the pre-selected risk event related description item;

[0107] A credibility determination unit, used to determine the credibility of the pre-selected risk event associated description items according to the extraction source;

[0108] The first item to be classified determining unit is used to use the pre-selected risk event associated description item of the corresponding extraction source as the risk event associated description item if the credibility is greater than or equal to a preset credibility threshold, and determine the first item to be classified according to the risk event associated description item.

[0109] The working principle and beneficial effects of the above technical solution are:

[0110] When obtaining risk events and event descriptions corresponding to different enterprises and enterprises through big data, there are situations where data needs to be screened. For example, if the description of the risk event is not appropriate, the suitability of the corresponding data for the subsequent plan formulation is low. Therefore, the risk event is analyzed to obtain the pre-selected first item to be classified. The pre-selected first item to be classified includes the enterprise risk event and the pre-selected risk event related description item corresponding to the enterprise risk event. The pre-selected risk event related description item is, for example, an analysis and evaluation of the enterprise risk event by a certain news platform, and another example: a risk event report released to the outside world by the enterprise corresponding to the enterprise risk event; determine the extraction source, the extraction source is the acquisition platform of the pre-selected risk event related description item, such as: a certain news platform, and another example: the medium of the risk event report released to the outside world by the enterprise corresponding to the enterprise risk event; when the credibility of the pre-selected risk event related description item is greater than or equal to the preset credibility threshold (the credibility threshold is preset manually, such as: 90), the pre-selected risk event related description item of the corresponding extraction source is used as the risk event related description item, and the first item to be classified is determined according to the risk event related description item, thereby improving the credibility of the first item to be classified.

[0111] In one embodiment, the credibility determination unit includes:

[0112] A first event attribute acquisition subunit, used to acquire a first event attribute of an enterprise risk event;

[0113] A second event attribute acquisition subunit, used to acquire a second event attribute of an extraction source;

[0114] A first credibility determination value determination subunit, configured to calculate the event attribute matching degree according to the first event attribute and the second event attribute, and use the matching degree as the first credibility determination value;

[0115] A correlation information calculation subunit, used to calculate the correlation degree and correlation direction according to the extraction source and the event party of the enterprise risk event;

[0116] An adjustment factor determination subunit, used to determine an adjustment factor of the first credibility determination value according to the correlation degree and the correlation direction;

[0117] A second credibility determination value determination subunit, configured to determine a second credibility determination value according to the adjustment factor and the first credibility determination value;

[0118] A credibility check value acquisition subunit is used to acquire a credibility check value of an extraction source;

[0119] A correction coefficient determination subunit, used to determine a correction coefficient of a second credibility determination value according to the credibility check value and the average credibility check value;

[0120] The credibility acquisition subunit is used to correct the second credibility determination value according to the correction coefficient to obtain the credibility of the extraction source corresponding to the pre-selected risk event related description item.

[0121] The working principle and beneficial effects of the above technical solution are:

[0122] The first event attribute is the attribute information of the enterprise risk event, such as: the industry to which the enterprise risk event belongs, the event type and the event cause, etc.; the second event attribute is determined according to the historical description information of the extraction source, specifically, the attribute information describing the event determined by parsing the historical description information; the first event attribute and the second event attribute are matched to determine the event attribute matching degree, and the event attribute matching degree represents the suitability of the extraction source for the description of the corresponding enterprise risk event. The higher the event attribute matching degree, the more credible the evaluation of the enterprise risk event by the corresponding extraction source; the event party of the enterprise risk event is: the enterprise where the enterprise risk event occurs; the correlation degree indicates the closeness of the relationship between the extraction source and the event party, which is determined based on the historical interaction behavior between the extraction source and the event party; the correlation direction includes positive correlation and negative correlation, and through specific analysis of historical interaction behavior, Analyze the interactive sentiment trend to determine; determine the adjustment factor of the first credibility determination value based on the correlation and correlation direction: if the correlation direction is positive correlation, when the correlation is greater than or equal to the preset correlation threshold, the adjustment factor is less than 1, for example: 0.8; when the correlation is less than the correlation threshold, the adjustment factor is equal to 1; when the correlation direction is negative correlation, obtain the unit distance of the correlation, and use the unit distance as the adjustment factor; when determining the second credibility determination value based on the adjustment factor and the first credibility determination value, multiply the adjustment factor and the first credibility determination value; the credibility verification value of the extraction source is the credit rating of the extraction source. When obtaining, obtain the credit rating of the corresponding extraction source according to the credit rating platform; the correction coefficient is the result obtained by dividing the credibility verification value by the average credibility verification value, and the correction coefficient is multiplied by the second credibility determination value to obtain the credibility.

[0123] The present invention determines the suitability of the description of the risk event associated description item by the extraction source by matching the first event attribute and the second event attribute, and uses it as the first credibility determination value; obtains the correlation and correlation direction between the extraction source and the event party according to the historical interaction behavior of the extraction source and the event party of the enterprise risk event, determines an adjustment factor according to the correlation and the correlation direction, adjusts the first credibility determination value according to the adjustment factor to obtain the second credibility determination value; obtains the credibility verification value of the extraction source, determines the correction coefficient according to the credibility verification value and the average credibility verification value, and corrects the second credibility determination value to obtain the credibility of the pre-selected risk event associated description item, and the credibility calculation accuracy is higher.

[0124] In one embodiment, the risk management subsystem includes:

[0125] An investment behavior dynamic determination module is used to dynamically determine investment behavior based on risk identification results;

[0126] Among them, the investment behavior dynamic determination module includes:

[0127] The rectification plan acquisition submodule is used to obtain the rectification plan of the investment enterprise based on the risk identification results;

[0128] The investment behavior determination submodule is used to monitor the rectification results of the rectification plan and determine the investment behavior.

[0129] The working principle and beneficial effects of the above technical solution are:

[0130] According to the risk identification results, the rectification plan of the investment enterprise is obtained: according to the risk identification results, the risk projects of the risk enterprise are determined, and the corresponding risk enterprise is notified to make rectifications; the subsequent management enterprise tracks and manages the risk enterprise to obtain the rectification results, and determines whether the risk enterprise has eliminated the corresponding risks. If so, the corresponding investment can continue, otherwise, the corresponding risk enterprise needs to be eliminated to improve the investment security of the enterprise.

[0131] In one embodiment, the rectification plan acquisition submodule includes:

[0132] A historical rectification enterprise determination unit is used to determine, based on the risk identification result, historical rectification enterprises whose rectification effect is greater than or equal to a preset rectification effect threshold; wherein the preset rectification effect threshold is manually preset; historical rectification enterprises are: enterprises that have experienced the risk corresponding to the risk identification result;

[0133] A historical rectification plan acquisition unit is used to acquire the historical rectification plan of the historical rectification enterprise corresponding to the risk identification result; wherein the historical rectification plan is: the rectification plan of the historical rectification enterprise for the risk identification result;

[0134] A first enterprise characteristic matrix acquisition unit is used to acquire a first enterprise characteristic matrix of a historically rectified enterprise; wherein the first enterprise characteristic matrix is ​​constructed based on the enterprise attribute characteristics of the historically rectified enterprise and is a description matrix of the historically rectified enterprise, and which matrix element position of the description matrix is ​​written into which enterprise attribute characteristic is manually preset;

[0135] A first local correlation matrix determination unit is used to determine a first local correlation matrix in a first enterprise feature matrix according to a historical rectification plan; wherein the first local correlation matrix is ​​an influence feature matrix of the historical rectification plan. When determining the first local correlation matrix, the influence features of the historical rectification plan in the first enterprise feature matrix (e.g., enterprise rectification resources) are determined, and a local matrix of the first enterprise feature matrix containing the influence features of all historical rectification plans is determined as the first local correlation matrix;

[0136] A second enterprise characteristic matrix determining unit, used to determine a second enterprise characteristic matrix of the venture capital enterprise; wherein the construction principle of the second enterprise characteristic matrix is ​​the same as the first enterprise characteristic matrix;

[0137] A second local correlation matrix determining unit, used to determine a second local correlation matrix in the second enterprise characteristic matrix according to the relative position of the first local correlation matrix and the first enterprise characteristic matrix; wherein the relative position is the matrix position of the first local correlation matrix relative to the first enterprise characteristic matrix;

[0138] The first matrix similarity calculation unit is used to calculate the first matrix similarity of the first local correlation matrix and the second local correlation matrix; the first matrix similarity Among them, a ij is the i-th row and j-th column element of the first local correlation matrix, b ij is the i-th row and j-th column element of the second local correlation matrix, m is the total number of rows in the matrix, and n is the total number of columns in the matrix;

[0139] The second matrix similarity calculation unit is used to calculate the second matrix similarity of the first local matrix of the first enterprise feature matrix excluding the first local correlation matrix and the second local matrix of the second enterprise feature matrix excluding the second local correlation matrix if the first matrix similarity is greater than or equal to a preset first matrix similarity threshold; wherein the preset first matrix similarity threshold is manually set, such as: 0.95; the enterprise feature matrix excluding the local correlation matrix refers to the enterprise feature matrix corresponding to the local correlation matrix The matrix element is set to 0;

[0140] A training rectification plan determination unit is used to use the corresponding historical rectification plan as the training rectification plan if the second matrix similarity is greater than or equal to a preset second matrix similarity threshold; wherein the first matrix similarity threshold is greater than the second matrix similarity threshold; the calculation rule of the second matrix similarity is the same as the first matrix similarity; the preset second matrix similarity threshold is manually pre-set, and when setting, the first matrix similarity threshold is greater than the second matrix similarity threshold, for example: 0.8;

[0141] The rectification plan acquisition unit is used to train the risk rectification model corresponding to the venture capital enterprise according to the training rectification plan, input the risk identification result into the risk rectification model, and obtain the rectification plan. Among them, the risk rectification model is: an AI model trained based on the training rectification plan, which automatically generates a rectification plan suitable for the venture capital enterprise according to the input risk.

[0142] The working principle and beneficial effects of the above technical solution are:

[0143] The present invention determines the historical rectification enterprises that have experienced the risks corresponding to the risk identification results, and obtains the rectification plans of the historical rectification enterprises for the risk identification results. Since the rectification plan is generally formulated based on the adaptability of enterprise attributes, it is necessary to conduct a rectification plan for the suitability analysis of subsequent deep learning.

[0144] Specifically, a first enterprise feature matrix is ​​constructed according to the enterprise attribute characteristics of the historical rectification enterprises, and a first local correlation matrix of the first enterprise feature matrix containing the influencing characteristics of the corresponding historical rectification plans is determined; based on the same construction principle of the first enterprise feature matrix, a second enterprise feature matrix is ​​constructed; according to the relative positions of the first local correlation matrix and the first enterprise feature matrix, a second local correlation matrix in the second enterprise feature matrix is ​​determined; the first matrix similarity of the first local correlation matrix and the second local correlation matrix is ​​calculated, and the corresponding first enterprise feature matrix and second enterprise feature matrix whose first matrix similarity is greater than or equal to a preset first matrix similarity threshold are screened for subsequent second matrix similarity calculation of the local matrix, and the historical rectification plans whose second matrix similarity is greater than or equal to the preset second matrix similarity threshold are used as training rectification plans. Since the suitability of the influencing enterprise characteristics of the historical rectification plans has a greater impact on the subsequent model training, the first matrix similarity threshold is set to be greater than the second matrix similarity threshold; the risk rectification model corresponding to the venture capital enterprise is trained according to the training rectification plan, and the rectification plan is automatically obtained based on the input risk, so that the rectification plan is more applicable.

[0145] The embodiment of the present invention provides a risk identification and management method for investment data, such as Figure 2 As shown, including:

[0146] Step 1: Obtain investment data of the management enterprise;

[0147] Step 2: Analyze investment data and obtain corporate portraits of invested enterprises;

[0148] Step 3: Obtain risk identification solution library;

[0149] Step 4: Determine the risk identification template corresponding to the investment enterprise based on the enterprise profile and risk identification solution library;

[0150] Step 5: Obtain risk identification results based on investment data and against the risk identification template;

[0151] Step 6: Conduct risk management for the investment enterprise based on the risk identification results.

[0152] In one embodiment, parsing investment data to obtain a corporate profile of an invested enterprise includes:

[0153] Get the enterprise portrait description template;

[0154] Obtain the corporate profile of the invested enterprise based on the corporate profile description template and investment data.

[0155] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include these modifications and variations.

Claims

1. A risk identification and management system for investment data, characterized in that: include: The investment data acquisition subsystem is used to acquire the investment data of the management enterprise; The enterprise portrait acquisition subsystem is used to analyze investment data and obtain the enterprise portrait of the invested enterprise; The risk identification solution library construction subsystem is used to obtain the risk identification solution library; The risk identification template determination subsystem is used to determine the risk identification template corresponding to the investment enterprise based on the enterprise profile and risk identification solution library; The risk identification result acquisition subsystem is used to obtain risk identification results based on investment data and risk identification templates; The risk management subsystem is used to conduct risk management on investment enterprises based on risk identification results.

2. A risk identification and management system for investment data according to claim 1, characterized in that: The enterprise portrait acquisition subsystem includes: An enterprise portrait description template acquisition module is used to obtain an enterprise portrait description template; The enterprise portrait acquisition module is used to obtain the enterprise portrait of the invested enterprise based on the enterprise portrait description template and investment data.

3. A risk identification and management system for investment data according to claim 1, characterized in that: The risk identification solution library construction subsystem includes: Enterprise risk event acquisition module, used to acquire enterprise risk events based on target big data nodes; An event analysis module is used to analyze enterprise risk events and obtain enterprise descriptions of risky enterprises and risk event-related description items; The risk identification scheme determination module is used to obtain the identification sub-strategies of the risk event associated description items, summarize the identification sub-strategies corresponding to the risk event associated description items corresponding to the same risk event, and obtain the risk identification scheme corresponding to the enterprise description; The module for determining items to be entered is used to associate the enterprise description with the risk identification plan to obtain items to be entered; The construction module is used to enter all items to be entered into the database and obtain a risk identification solution library.

4. A risk identification and management system for investment data as claimed in claim 3, characterized in that: Event parsing module, including: A first item to be classified obtaining submodule, used for analyzing enterprise risk events and obtaining a first item to be classified; The first item to be classified is: a one-to-one corresponding description of the enterprise to be classified and a description item associated with the first risk event to be classified; The second item to be classified acquisition submodule is used to acquire the second item to be classified according to the first item to be classified based on the first classification rule corresponding to the enterprise description; the second item to be classified is: a one-to-one corresponding enterprise description and a description item associated with the second risk event to be classified; The classification result acquisition submodule is used to obtain the classification result based on the second classification rule corresponding to the enterprise risk event and the second item to be classified; the classification result includes: one-to-one corresponding enterprise description, risk event identifier and risk event related description item corresponding to the risk event identifier.

5. A risk identification and management system for investment data as claimed in claim 4, characterized in that: The first to-be-classified item acquisition submodule includes: A pre-selected first item to be classified obtaining unit is used to analyze the enterprise risk event and obtain the pre-selected first item to be classified; An extraction source acquisition unit, used to parse the first item to be classified and acquire the extraction source of the pre-selected risk event related description item; A credibility determination unit, used to determine the credibility of the pre-selected risk event associated description items according to the extraction source; The first item to be classified determining unit is used to use the pre-selected risk event associated description item of the corresponding extraction source as the risk event associated description item if the credibility is greater than or equal to a preset credibility threshold, and determine the first item to be classified according to the risk event associated description item.

6. A risk identification and management system for investment data according to claim 1, characterized in that: The risk identification result acquisition subsystem includes: The risk data acquisition module is used to obtain the risk data of the investment enterprise according to the identification basis of the risk identification template; The risk identification result acquisition module is used to obtain risk identification results based on risk data and risk identification templates.

7. A risk identification and management system for investment data according to claim 1, characterized in that: Risk management subsystem, including: The investment behavior dynamic determination module is used to dynamically determine the investment behavior based on the risk identification results.

8. A risk identification and management system for investment data as claimed in claim 7, characterized in that: The investment behavior dynamic determination module includes: The rectification plan acquisition submodule is used to obtain the rectification plan of the investment enterprise based on the risk identification results; The investment behavior determination submodule is used to monitor the rectification results of the rectification plan and determine the investment behavior.

9. A risk identification and management method for investment data, characterized in that: include: Obtain investment data of management enterprises; Analyze investment data and obtain corporate portraits of invested companies; Obtain a library of risk identification solutions; Determine the risk identification template corresponding to the investment enterprise based on the enterprise portrait and risk identification solution library; Obtain risk identification results based on investment data and risk identification templates; Conduct risk management for investment enterprises based on risk identification results.

10. A risk identification and management method for investment data according to claim 9, characterized in that: Analyze investment data and obtain corporate portraits of invested companies, including: Get the enterprise portrait description template; Obtain the corporate profile of the invested enterprise based on the corporate profile description template and investment data.

Images