Scenarized risk automatic assessment method, computing device and storage medium

Through a scenario-based risk automation assessment method, using the risk scenario library and evaluation model, the problems of poor accuracy and low efficiency of risk assessment in the existing technology are solved, and fast and accurate risk assessment is achieved, reducing human error and manpower investment .

CN119090266AInactive Publication Date: 2024-12-06BEIJING COMPLIANCE TECHNOLOGY CO LTD

Patent Information

Application Number
CN202411120506.6
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-08-15
Publication Date
2024-12-06
Estimated Expiration
Not applicable · inactive patent

AI Technical Summary

Technical Problem

The existing risk assessment methods are inaccurate and inefficient, relying on expert experience and limited data, resulting in inaccurate risk assessment and long time, and prone to artificial errors.

Method used

A scenario-based risk automation evaluation method is adopted. By obtaining the input data of the scenario to be evaluated, matching risk scenarios are selected from the preset risk scenario database, risk factors are determined, and risk assessment models are used to calculate the risk level and output the risk level. The method includes feature extraction, similarity calculation, Bayesian estimation and the use of weighted averages to improve the accuracy and efficiency of the evaluation.

Benefits of technology

A rapid and accurate risk assessment is achieved, manpower investment is reduced, man-made errors are reduced, and evaluation efficiency and accuracy are improved.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119090266A_ABST
    Figure CN119090266A_ABST
Patent Text Reader

Abstract

The invention provides a scenarized risk automatic assessment method, computing equipment and a storage medium. The method comprises the steps of obtaining input data of a to-be-evaluated scene; selecting at least one risk scene matched with the input data from a preset risk scene library; determining a risk factor of the to-be-evaluated scene according to the at least one risk scene; calculating a risk level of the to-be-evaluated scene according to the risk factor by using a risk evaluation model; and outputting the risk level. According to the technical scheme provided by the invention, the accuracy, efficiency and flexibility of risk assessment are improved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The embodiments of the present application relate to the field of network security technology, and in particular to a scenario-based automated risk assessment method, computing device, and storage medium. Background Art

[0002] With the expansion of business scale and the development of technology, traditional manual risk assessment methods are increasingly unable to meet the needs of modern enterprises. Enterprises need a method that can quickly and accurately assess risks so that they can take timely measures to reduce risks.

[0003] Most current risk assessment methods rely on expert experience or simple statistical analysis, which have the following shortcomings:

[0004] Lack of accuracy: Due to the lack of a systematic risk scenario library and an automatic identification mechanism for risk factors, existing risk assessment methods can only rely on expert experience and limited data, which may lead to inaccurate risk assessment.

[0005] Inefficiency: Traditional risk assessment usually requires a lot of manpower, which is not only time-consuming but also prone to human errors. Summary of the invention

[0006] The embodiments of the present application provide a scenario-based automated risk assessment method, computing device, and storage medium to solve the problems of poor accuracy and low efficiency in risk assessment in the prior art.

[0007] In a first aspect, an embodiment of the present application provides a scenario-based automated risk assessment method, including:

[0008] Obtain input data for the scenario to be evaluated;

[0009] Selecting at least one risk scenario matching the input data from a preset risk scenario library;

[0010] Determining, according to the at least one risk scenario, a risk factor of the scenario to be assessed;

[0011] Calculating the risk level of the scenario to be assessed based on the risk factors using a risk assessment model;

[0012] The risk level is output.

[0013] Optionally, the selecting at least one risk scenario matching the input data from a preset risk scenario library includes:

[0014] Performing feature extraction on the input data to obtain a feature vector;

[0015] In the risk scenario library, at least one risk scenario whose similarity to the feature vector satisfies a similarity condition is searched.

[0016] Optionally, the searching for at least one risk scenario whose similarity to the feature vector satisfies a similarity condition comprises:

[0017] Calculate the cosine similarity between the feature vector and the feature vector of each risk scenario, and select the risk scenario whose cosine similarity meets the similarity condition; or,

[0018] Calculating the cosine similarity between the feature vector and the feature vector of each risk scenario;

[0019] Calculating the Manhattan distance between the feature vector and the feature vector of the selected risk scenario, and calculating the degree of matching between the feature vector and the feature vector of the selected risk scenario based on the Manhattan distance;

[0020] Among them, through the formula: Calculate the matching degree between the feature vector and the feature vector of the selected risk scenario, S represents the matching degree between the feature vector and the feature vector of the selected risk scenario, α is an adjustment parameter used to balance the influence of cosine similarity and Manhattan distance; D represents Manhattan distance;

[0021] Using the weighted average of the cosine similarity and the matching degree as the comprehensive similarity, and screening out risk scenarios whose comprehensive similarities meet similarity conditions;

[0022] The comprehensive similarity is calculated by the formula: CS=β·C+(1-β)·S, where CS represents the comprehensive similarity, β is a preset weighting coefficient, C represents the cosine similarity, and S represents the matching degree.

[0023] Optionally, determining the risk factor of the scenario to be evaluated according to the at least one risk scenario includes:

[0024] identifying key variables in the at least one risk scenario;

[0025] Mapping the key variables to the scenario to be evaluated to obtain risk factors;

[0026] According to the distribution characteristics of risk factors, Bayesian estimation is used to update the probability distribution of risk factors.

[0027] Optionally, the using a risk assessment model to calculate the risk level of the scenario to be assessed according to the risk factor includes:

[0028] The risk factor is input into a pre-trained risk assessment model, and the risk level is output using the risk assessment model.

[0029] Optionally, the outputting the risk level using a risk assessment model includes:

[0030] The risk level is calculated using the following formula:

[0031]

[0032] Among them, R represents the risk level, f i represents the i-th risk factor, w i is with f i The corresponding weight coefficient, μ i and σ i represent the expected value and standard deviation of the ith risk factor respectively.

[0033] Optionally, the weight coefficient w i Determined by the following formula:

[0034]

[0035] Among them, a i is the preset adjustment coefficient, m i represents the ideal value and allowable range of the ith risk factor, r i represents the permissible range of the ith risk factor, i.e., f i With m i The allowable deviation between .

[0036] Optionally, the outputting the risk level includes:

[0037] The risk level is displayed on a user interface; or the risk level is sent to a remote server, so that the remote server performs corresponding processing based on the risk level.

[0038] In a second aspect, an embodiment of the present application provides a computing device, including a processing component and a storage component; the storage component stores one or more computer instructions; the one or more computer instructions are used to be called and executed by the processing component to implement the scenario-based automated risk assessment method as described in the first aspect above.

[0039] In a third aspect, an embodiment of the present application provides a computer storage medium storing a computer program, which, when executed by a computer, implements the scenario-based automated risk assessment method as described in the first aspect above.

[0040] In an embodiment of the present application, input data of a scenario to be evaluated is obtained; at least one risk scenario matching the input data is selected from a preset risk scenario library; a risk factor of the scenario to be evaluated is determined based on the at least one risk scenario; a risk assessment model is used to calculate the risk level of the scenario to be evaluated based on the risk factor; and the risk level is output.

[0041] These and other aspects of the present application will become more clearly understood in the description of the following embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, a brief introduction will be given below to the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0043] Figure 1 A flowchart of a scenario-based automated risk assessment method provided in an embodiment of the present application;

[0044] Figure 2 A schematic diagram of the structure of a scenario-based automated risk assessment device provided in an embodiment of the present application;

[0045] Figure 3 A schematic diagram of the structure of a computing device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0046] In order to enable those skilled in the art to better understand the solution of the present application, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application.

[0047] In some of the processes described in the specification and claims of this application and the above-mentioned figures, multiple operations that appear in a specific order are included, but it should be clearly understood that these operations may not be executed in the order in which they appear in this article or executed in parallel. The serial numbers of the operations, such as 101, 102, etc., are only used to distinguish between different operations, and the serial numbers themselves do not represent any execution order. In addition, these processes may include more or fewer operations, and these operations may be executed in sequence or in parallel. It should be noted that the descriptions of "first", "second", etc. in this article are used to distinguish different messages, devices, modules, etc., do not represent the order of precedence, and do not limit the "first" and "second" to be different types.

[0048] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of this application.

[0049] Figure 1 A flowchart of a scenario-based automated risk assessment method is provided for an embodiment of the present application, such as Figure 1 As shown, the method includes:

[0050] 101. Obtain input data of the scenario to be evaluated;

[0051] In this step, the input data of the scenario to be evaluated is obtained: the concepts involved in this step mainly include "scenario to be evaluated" and "input data".

[0052] Scenario to be assessed: refers to the specific situation or environment that needs to be assessed for risk. It can be any situation that needs to assess potential risks, such as financial transactions, cybersecurity incidents, production safety accidents, etc.

[0053] Input data: refers to various data and information used to assess the risks of a specific scenario. These data can be quantitative data (such as transaction amount, traffic size, etc.) or qualitative data (such as behavior patterns, security vulnerabilities, etc.).

[0054] In the embodiments of this application, we will take financial investment decisions as an example to illustrate how to obtain input data for the scenario to be evaluated.

[0055] Scenario to be evaluated: Assume that the scenario to be evaluated is a financial investment decision scenario, where investors need to evaluate the risk level of a certain investment project in order to make a reasonable investment decision.

[0056] Input data: Input data includes but is not limited to the following:

[0057] Market data: including stock prices, bond yields, foreign exchange rates, etc.

[0058] Macroeconomic data: such as GDP growth rate, inflation rate, etc.

[0059] Industry data: such as industry growth rate, competitor status, etc.

[0060] Company financial data: such as operating income, net profit, debt ratio, etc.

[0061] Policy and regulatory data: such as the latest financial regulatory policies, etc.

[0062] Specific example: Assume that an investor is evaluating the investment risk of a startup company and needs to collect the following data:

[0063] Market data: stock price fluctuations in the past year and the stock price performance of major competitors in the industry.

[0064] Macroeconomic data: current GDP growth rate, unemployment rate, etc.

[0065] Industry data: industry growth rate in the past five years, changes in market share of major competitors, etc.

[0066] Company financial data: financial statements for the past three years, including revenue, profit, cash flow, etc.

[0067] Policy and regulatory data: the latest industry regulatory policies and regulatory changes.

[0068] Data collection: Investors can obtain this data through a variety of channels, such as public financial reports, industry research reports, economic indicator release websites, etc.

[0069] Data preprocessing: The collected data may need to be cleaned and preprocessed to remove invalid data or missing values ​​and convert the data into a form suitable for subsequent analysis. For example, unstructured text data can be converted into a structured data format such as CSV or Excel table.

[0070] Application example: Investors collected financial data of a startup company, including operating income of 1 million yuan, 1.2 million yuan, and 1.5 million yuan in the past three years; net profit of 100,000 yuan, 150,000 yuan, and 200,000 yuan; debt ratio of 20%, 25%, and 30%. In addition, industry data was collected, and the industry growth rate was 5%, 6%, and 7%. Based on this data, investors can further analyze the financial health of the company and the growth trend of its industry.

[0071] 102. Select at least one risk scenario matching the input data from a preset risk scenario library;

[0072] In this step, the risk scenario library: assume that the risk scenario library contains multiple risk scenarios related to financial investment, and each scenario has its own feature vector, including but not limited to market volatility, economic indicators, industry growth rate, etc.

[0073] Input data: Investors collected financial data about a startup company, including operating income, net profit, debt ratio, etc. for the past three years.

[0074] Optionally, in the embodiment of the present application, step 102 may specifically include:

[0075] 1021. Perform feature extraction on the input data to obtain a feature vector;

[0076] In this step, feature extraction: extract key features from the input data to form a feature vector. For example, operating income, net profit, debt ratio, etc. can be used as components of the feature vector.

[0077] Feature vector: A data representation in the form of a vector consisting of key features extracted from the input data.

[0078] 1022. In the risk scenario library, search for at least one risk scenario whose similarity to the feature vector satisfies a similarity condition.

[0079] The similarity condition may be satisfying a preset threshold.

[0080] In some embodiments, step 1022 may specifically include: calculating the cosine similarity between the feature vector and the feature vector of each risk scenario, and selecting the risk scenario whose cosine similarity meets the similarity condition.

[0081] In this step, cosine similarity: a measure of the directional similarity of two vectors, with a value range between -1 and 1. The closer the value is to 1, the more similar the two vectors are.

[0082] In the embodiment of this application, it is assumed that the investor has collected the financial data of a startup company, including the operating income of the past three years, which were 1 million yuan, 1.2 million yuan, and 1.5 million yuan respectively; the net profit was 100,000 yuan, 150,000 yuan, and 200,000 yuan respectively; the debt ratio was 20%, 25%, and 30%. The risk scenario library contains three risk scenarios, and the feature vector of each risk scenario is as follows:

[0083] Risk scenario 1: operating income = 1 million yuan, net profit = 100,000 yuan, debt ratio = 20%;

[0084] Risk scenario 2: operating income = 1.2 million yuan, net profit = 150,000 yuan, debt ratio = 25%;

[0085] Risk scenario 3: operating income = 1.5 million yuan, net profit = 200,000 yuan, debt ratio = 30%;

[0086] Convert the collected financial data into feature vector form, for example:

[0087] income_vector = [100,120,150];

[0088] Net profit vector = [10, 15, 20];

[0089] Debt ratio vector = [20, 25, 30];

[0090] Calculate the cosine similarity between the feature vector of the input data and the feature vector of each risk scenario in the risk scenario library. For example, for risk scenario 1:

[0091] Cosine similarity C1 = cos(income vector, income vector of risk scenario 1) × cos(net profit vector, net profit vector of risk scenario 1) × cos(debt ratio vector, debt ratio vector of risk scenario 1).

[0092] Further, it is determined whether the cosine similarity C1 meets a preset threshold, and if so, the risk scenario 1 is used as at least one risk scenario that matches the input data. Similarly, the risk scenarios 2 and 3 are determined.

[0093] In some other embodiments, step 1022 may specifically include: calculating the cosine similarity between the feature vector and the feature vector of each risk scenario; calculating the Manhattan distance between the feature vector and the feature vector of the selected risk scenario, and calculating the degree of matching between the feature vector and the feature vector of the selected risk scenario based on the Manhattan distance; using the weighted average of the cosine similarity and the degree of matching as the comprehensive similarity, and screening out the risk scenarios whose comprehensive similarities meet the similarity conditions;

[0094] Among them, through the formula: Calculate the matching degree between the feature vector and the feature vector of the selected risk scenario, S represents the matching degree between the feature vector and the feature vector of the selected risk scenario, α is an adjustment parameter used to balance the influence of cosine similarity and Manhattan distance; D represents Manhattan distance;

[0095] The comprehensive similarity is calculated by the formula: CS=β·C+(1-β)·S, where CS represents the comprehensive similarity, β is a preset weighting coefficient, C represents the cosine similarity, and S represents the matching degree.

[0096] In this step, Manhattan distance calculation: The Manhattan distance between the feature vector of the input data and the feature vector of the selected risk scenario is calculated.

[0097] Matching degree calculation: Calculate the matching degree S based on the Manhattan distance.

[0098] Comprehensive similarity calculation: The weighted average of cosine similarity and matching degree is used as the comprehensive similarity CSCS.

[0099] Risk scenario selection: Filter out risk scenarios that match the input data based on comprehensive similarity.

[0100] In an embodiment of the present application, continuing with the above example, after obtaining the cosine similarities of risk scenario 1, risk scenario 2, and risk scenario 3, the Manhattan distance between the feature vector and the feature vector of the selected risk scenario is calculated.

[0101] Specifically, the Manhattan distance between the feature vector of the input data and the feature vector of the selected risk scenario is calculated. For example, for risk scenario 1: Manhattan distance D1 = |100-100| + |10-10| + |20-20|;

[0102] Furthermore, the matching degree S is calculated according to the Manhattan distance. For example, for risk scenario 1, assuming α = 0.1:

[0103] Matching degree

[0104] The weighted average of cosine similarity and matching degree is used as the comprehensive similarity CS. For example, assuming β = 0.7:

[0105] Comprehensive similarity CS1 = 0.7 × C1 + 0.3 × S1;

[0106] Here is a specific example:

[0107] income_vector = [100,120,150];

[0108] Net profit vector = [10, 15, 20];

[0109] Debt ratio vector = [20, 25, 30];

[0110] Risk scenario 1: Operating income = 1 million yuan, net profit = 100,000 yuan, debt ratio = 20%

[0111] Cosine similarity C1 = cos([100,120,150],[100,10,20])×cos([10,15,20],[10,15,20])×cos([20,25,30],[20,25,30]);

[0112] Manhattan distance D1 = |100-100| + |10-10| + |20-20| = 0;

[0113] Matching degree S1 = 1 / (1 + 0.1 × 0) = 1;

[0114] Comprehensive similarity CS1 = 0.7 × C_1 + 0.3 × 1;

[0115] Calculation results:

[0116] Cosine similarity: Since the calculation of cosine similarity involves the calculation of the angle between vectors, we simplify the process here and assume that C1 = 0.9.

[0117] Comprehensive similarity: Comprehensive similarity CS1 = 0.7 × 0.9 + 0.3 × 1 = 0.93;

[0118] Through the above steps, investors can find risk scenarios that match the input data and better assess potential risks.

[0119] 103. Determine a risk factor of the scenario to be evaluated according to the at least one risk scenario;

[0120] Optionally, in the embodiment of the present application, step 103 may specifically include:

[0121] 1031. Identify key variables in the at least one risk scenario;

[0122] 1032. Map the key variables to the scenario to be evaluated to obtain risk factors;

[0123] 1033. Based on the distribution characteristics of risk factors, Bayesian estimation is used to update the probability distribution of risk factors.

[0124] In this step, the risk factor of the scenario to be evaluated is determined based on the at least one risk scenario: the concepts involved in this step mainly include "risk scenario", "key variables", "risk factor" and "Bayesian estimation".

[0125] Risk scenario: refers to one or more risk scenarios selected in the previous steps that match the scenario to be evaluated.

[0126] Key variables: refers to factors or indicators that have a significant impact on risks in risk scenarios.

[0127] Risk factor: refers to the specific factors or indicators mapped to the scenario to be evaluated based on key variables, which are used to assess the risk level of the scenario to be evaluated.

[0128] Bayesian estimation: A statistical method for updating the probability distribution of a parameter based on prior knowledge and observed data.

[0129] In the embodiment of the present application, taking financial investment decision as an example, it is explained how to determine the risk factor of the scenario to be evaluated based on at least one risk scenario.

[0130] Risk scenarios: Assume that three risk scenarios matching the input data are selected from the risk scenario library.

[0131] Identification of key variables: Identify key variables that have a significant impact on the risks from these three risk scenarios.

[0132] Risk factor mapping: Map these key variables to the scenario to be evaluated to obtain the risk factors of the scenario to be evaluated.

[0133] Probability distribution update: Based on the distribution characteristics of risk factors, the probability distribution of risk factors is updated using Bayesian estimation.

[0134] Here is a specific example:

[0135] Assume that an investor has collected financial data of a startup company, including operating income of RMB 1 million, RMB 1.2 million, and RMB 1.5 million in the past three years; net profit of RMB 100,000, RMB 150,000, and RMB 200,000; debt ratio of 20%, 25%, and 30%. Three risk scenarios are selected from the risk scenario library, and the key variables of each risk scenario are as follows:

[0136] Risk scenario 1: Changes in operating income growth rate, net profit growth rate, and debt ratio;

[0137] Risk scenario 2: industry growth rate, macroeconomic indicators, debt ratio;

[0138] Risk scenario 3: market volatility, competition situation, net profit growth rate;

[0139] Identification of key variables:

[0140] Risk scenario 1: operating income growth rate = 20%, net profit growth rate = 50%, debt ratio change = 5%;

[0141] Risk scenario 2: Industry growth rate = 5%, macroeconomic indicators = 2%, debt ratio = 25%;

[0142] Risk scenario 3: Market volatility = 10%, competition = medium, net profit growth rate = 50%;

[0143] Risk Factor Mapping:

[0144] Operating income growth rate: The operating income growth rate mapped to the scenario to be evaluated.

[0145] Net profit growth rate: the net profit growth rate mapped to the scenario to be evaluated.

[0146] Debt ratio change: Maps to the debt ratio change in the scenario to be evaluated.

[0147] Industry growth rate: The industry growth rate mapped to the scenario to be evaluated.

[0148] Macroeconomic indicators: Macroeconomic indicators mapped to the scenario to be evaluated.

[0149] Debt ratio: Maps to the debt ratio in the scenario to be evaluated.

[0150] Market volatility: The market volatility mapped to the scenario to be evaluated.

[0151] Competitive Situation: Map the competitive situation to the scenario to be evaluated.

[0152] Probability distribution update:

[0153] Bayesian estimation: Assume that the prior distribution of the operating income growth rate in the scenario to be evaluated is a normal distribution Where μ1 is the mean and σ1 is the standard deviation. Use Bayesian estimation to update the probability distribution of operating income growth rate.

[0154] Example calculation:

[0155] Operating income growth rate: The operating income in the past three years was RMB 1 million, RMB 1.2 million, and RMB 1.5 million respectively. Calculate the operating income growth rate.

[0156] Net profit growth rate: The net profits in the past three years were RMB 100,000, RMB 150,000, and RMB 200,000 respectively. Calculate the net profit growth rate.

[0157] Change in debt ratio: The debt ratio in the past three years was 20%, 25% and 30% respectively. Calculate the change in debt ratio.

[0158] Industry growth rate: Assume that the average growth rate of the industry over the past five years is 5%.

[0159] Macroeconomic indicators: Assume that the current macroeconomic indicator is 2%.

[0160] Debt ratio: The debt ratios in the past three years were 20%, 25% and 30% respectively.

[0161] Market Volatility: Assume the market volatility over the past year was 10%.

[0162] Competitive situation: Assume that the competitive situation is moderate.

[0163] Example calculation:

[0164] Operating income growth rate: The average value is 0.225.

[0165] Net profit growth rate: The average value is 0.415.

[0166] Changes in debt ratio: The average value is 0.225.

[0167] Probability distribution update:

[0168] Assume that the prior distribution of operating income growth rate is N(0.2,0.1 2 ).

[0169] Operating income growth rate: Assuming the observed operating income growth rate is 0.225, use Bayesian estimation to update the probability distribution of the operating income growth rate.

[0170] Bayesian estimation formula: The updated distribution is in,

[0171] n1 is the sample size in the prior distribution (usually set to 1), n2 is the observed sample size (set to 1 here), is the observed value, σ x is the standard deviation of the observations (here assumed to be 0.1).

[0172] Calculation process:

[0173] n1=1, n2=1, μ1=0.2, σ1=0.1, σ x =0.1

[0174]

[0175] Therefore, the updated distribution is N(0.2125,0.0333).

[0176] Through the above steps, investors can determine the risk factors of the scenario to be evaluated based on the key variables in the risk scenario, and use Bayesian estimation to update the probability distribution of the risk factors, so as to better assess potential risks.

[0177] 104. Calculate the risk level of the scenario to be evaluated according to the risk factors using a risk assessment model;

[0178] Optionally, in an embodiment of the present application, step 104 may specifically include: inputting the risk factor into a pre-trained risk assessment model, and using the risk assessment model to output the risk level.

[0179] Wherein, the outputting of the risk level by using the risk assessment model includes:

[0180] The risk level is calculated using the following formula:

[0181]

[0182] Among them, R represents the risk level, f i represents the i-th risk factor, w i is with f i The corresponding weight coefficient, μ i and σi represent the expected value and standard deviation of the ith risk factor respectively.

[0183] Among them, the weight coefficient w i Determined by the following formula:

[0184]

[0185] Among them, a i is the preset adjustment coefficient, m i represents the ideal value and allowable range of the ith risk factor, r i represents the permissible range of the ith risk factor, i.e., f i With m i The allowable deviation between .

[0186] In this step, the risk assessment model is used to calculate the risk level of the scenario to be assessed based on the risk factors: the concepts involved in this step mainly include "risk assessment model", "risk factor", "risk level", "weight coefficient" and "expected value and standard deviation".

[0187] Risk Assessment Model: A pre-trained model that predicts the risk level based on the input risk factors.

[0188] Risk factors: Specific risk factors in the scenario to be evaluated determined based on the risk scenario.

[0189] Risk level: A quantitative indicator that assesses the risk level of the scenario to be assessed.

[0190] Weight coefficient: used to adjust the importance of different risk factors when calculating risk levels.

[0191] Expected value and standard deviation: statistical properties of risk factors, used to measure the central location and dispersion of risk factors.

[0192] In the embodiment of the present application, a risk assessment model is used to calculate the risk level of the scenario to be assessed based on the risk factors. For example, this embodiment takes financial investment decisions as an example to illustrate how to calculate the risk level of the scenario to be assessed based on the risk factors.

[0193] Risk factors: Assume that investors have determined the risk factors of the scenario to be evaluated based on the previous steps, including operating income growth rate, net profit growth rate, debt ratio changes, etc.

[0194] Risk Assessment Model: Use a pre-trained risk assessment model to calculate the risk level.

[0195] Risk level calculation: Calculate the risk level of the scenario to be evaluated based on risk factors.

[0196] Weight coefficient calculation: Calculate the weight coefficient of each risk factor.

[0197] Expected Value and Standard Deviation: Set the expected value and standard deviation for each risk factor.

[0198] Example calculation:

[0199] Assume that an investor has collected financial data of a startup company, including operating income of RMB 1 million, RMB 1.2 million, and RMB 1.5 million in the past three years; net profit of RMB 100,000, RMB 150,000, and RMB 200,000; debt ratio of 20%, 25%, and 30%. Three risk scenarios are selected from the risk scenario library, and the following risk factors are determined:

[0200] Operating income growth rate: The operating income growth rates in the past three years were 0.2, 0.25, and 0.225, respectively, with an average of 0.225.

[0201] Net profit growth rate: The net profit growth rates in the past three years were 0.5, 0.33 and 0.415 respectively, with an average of 0.415.

[0202] Change in debt ratio: The changes in debt ratio over the past three years were 0.25, 0.2, and 0.225, respectively, with an average of 0.225.

[0203] Risk factors:

[0204] Operating income growth rate: f1 = 0.225;

[0205] Net profit growth rate: f2 = 0.415;

[0206] Change in debt ratio: f3 = 0.225;

[0207] Expected value and standard deviation:

[0208] Operating income growth rate: μ1 = 0.2125, σ1 = 0.0333;

[0209] Net profit growth rate: μ2 = 0.415, σ2 = 0.0725;

[0210] Debt ratio change: μ3 = 0.225, σ3 = 0.025;

[0211] Weight coefficient calculation:

[0212] Adjustment coefficient: Assume a1 = 0.5, a2 = 0.8, a3 = 0.7;

[0213] Ideal value: Assume m1 = 0.2, m2 = 0.4, m3 = 0.2;

[0214] Allowable range: Assume r1 = 0.05, r2 = 0.1, r3 = 0.05;

[0215] Weight coefficient calculation formula:

[0216]

[0217] Weight coefficient calculation example:

[0218] Calculate w1:

[0219] Calculate w2:

[0220] Calculate w3:

[0221] Risk level calculation:

[0222] Risk level calculation formula:

[0223]

[0224] Risk level calculation example:

[0225] Calculate R:

[0226] Substituting w1, w2, and w3 into the above, we get:

[0227]

[0228] Through the above steps, investors can calculate the risk level of the scenario to be evaluated based on the risk factors, so as to better assess the potential risks.

[0229] 105. Output the risk level.

[0230] Optionally, in an embodiment of the present application, step 105 may specifically include: displaying the risk level on a user interface; or sending the risk level to a remote server, so that the remote server performs corresponding processing based on the risk level.

[0231] In this step, the risk level is output: the concepts involved in this step mainly include "risk level", "user interface" and "remote server".

[0232] Risk level: refers to the risk level of the scenario to be assessed calculated by the risk assessment model.

[0233] User interface: The interface through which the user interacts with the system, which can be a graphical user interface (GUI), command line interface, or any other form of user interaction interface.

[0234] Remote Server: A server located at a remote location that receives risk level information and may perform further processing or decisions based on that information.

[0235] In the embodiment of the present application, taking financial investment decision as an example, it is explained how to output the risk level of the scenario to be evaluated.

[0236] Risk level: Assume that the risk level calculated in the previous steps is RR.

[0237] User Interface: User interface showing risk levels.

[0238] Remote server: A remote server that receives risk levels and performs corresponding processing.

[0239] Assume that the investor has calculated the risk level RR of the scenario to be evaluated according to the previous steps. Now, the investor needs to output the risk level for the next decision or processing.

[0240] User interface output:

[0241] User Interface: The interface of an application or system used by investors.

[0242] Output risk level: Display the risk level RR on the user interface for investors to view.

[0243] An investor opens an application, and the value of the risk level RR is displayed on the interface of the application, for example, R=0.75.

[0244] Remote server output:

[0245] Remote server: Assume that the financial institution where the investor works has a risk management platform that is located on a remote server.

[0246] Send risk level: Send the risk level R to the remote server so that the platform can perform further processing based on the risk level.

[0247] Embodiment: The investor's application sends a risk level R=0.75 to a remote server. After receiving the risk level, the remote server performs corresponding processing based on the risk level, such as triggering an alarm, generating a report, or adjusting the risk management strategy.

[0248] Example implementation:

[0249] Assume that the investor calculates the risk level R = 0.75 according to the previous steps.

[0250] User interface output example:

[0251] User Interface: The application interface used by investors.

[0252] Output risk level: The risk level R=0.75 is displayed on the application interface.

[0253] Example: When an investor opens the application, the following information is displayed on the interface: "Risk Level: 0.75". The investor can make corresponding decisions based on this information.

[0254] Example of remote server output:

[0255] Remote server: risk management platform of the financial institution where the investor is located.

[0256] Sending risk level: The investor's application sends the risk level R=0.75 to the remote server.

[0257] Embodiment: The investor's application sends the risk level R=0.75 to the remote server via the network. After receiving the risk level, the remote server may trigger a corresponding alarm or adjust the risk management strategy, such as increasing the monitoring frequency, adjusting the investment portfolio, etc.

[0258] Through the above steps, investors can not only intuitively understand the risk level of the scenario to be evaluated, but also conduct further processing or decision-making through remote servers, so as to better manage and respond to potential risks.

[0259] Figure 2 A schematic diagram of the structure of a scenario-based automated risk assessment device is provided for an embodiment of the present application, such as Figure 2 As shown, the device comprises:

[0260] An acquisition module 21 is used to acquire input data of the scenario to be evaluated;

[0261] A selection module 22, configured to select at least one risk scenario matching the input data from a preset risk scenario library;

[0262] A determination module 23, configured to determine a risk factor of the scenario to be evaluated according to the at least one risk scenario;

[0263] A calculation module 24, configured to calculate the risk level of the scenario to be evaluated according to the risk factors using a risk assessment model;

[0264] The output module 25 is used to output the risk level.

[0265] Optionally, in an embodiment of the present application, the selection module 22 is specifically used to extract features from the input data to obtain a feature vector; and in the risk scenario library, to search for at least one risk scenario whose similarity to the feature vector satisfies a similarity condition.

[0266] Optionally, in the embodiment of the present application, the selection module 22 is further used to calculate the cosine similarity between the feature vector and the feature vector of each risk scenario, and select the risk scenario whose cosine similarity satisfies the similarity condition; or, calculate the cosine similarity between the feature vector and the feature vector of each risk scenario; calculate the Manhattan distance between the feature vector and the feature vector of the selected risk scenario, and calculate the matching degree between the feature vector and the feature vector of the selected risk scenario based on the Manhattan distance; wherein, by the formula: Calculate the degree of match between the feature vector and the feature vector of the selected risk scenario, where S represents the degree of match between the feature vector and the feature vector of the selected risk scenario, α is an adjustment parameter used to balance the influence of cosine similarity and Manhattan distance; D represents Manhattan distance; use the weighted average of cosine similarity and the degree of match as the comprehensive similarity, and screen out risk scenarios whose comprehensive similarities meet similarity conditions; wherein the comprehensive similarity is calculated by the formula: CS=β·C+(1-β)·S, CS represents the comprehensive similarity, β is a preset weighting coefficient, C represents the cosine similarity, and S represents the degree of match.

[0267] Optionally, in an embodiment of the present application, the determination module 23 is specifically used to identify key variables in the at least one risk scenario; map the key variables to the scenario to be evaluated to obtain risk factors; and update the probability distribution of the risk factors using Bayesian estimation based on the distribution characteristics of the risk factors.

[0268] Optionally, in the embodiment of the present application, the calculation module 24 is specifically used to input the risk factor into a pre-trained risk assessment model, and output the risk level using the risk assessment model;

[0269] Wherein, the outputting of the risk level by using the risk assessment model includes:

[0270] The risk level is calculated using the following formula:

[0271]

[0272] Among them, R represents the risk level, f i represents the i-th risk factor, w i is with f i The corresponding weight coefficient, μ i and σ i represent the expected value and standard deviation of the ith risk factor respectively.

[0273] Among them, the weight coefficient w i Determined by the following formula:

[0274]

[0275] Among them, a i is the preset adjustment coefficient, m i represents the ideal value and allowable range of the ith risk factor, r i represents the permissible range of the ith risk factor, i.e., f i With m i The allowable deviation between .

[0276] Optionally, in an embodiment of the present application, the output module 25 is specifically used to display the risk level on a user interface; or send the risk level to a remote server so that the remote server performs corresponding processing based on the risk level.

[0277] Figure 2 The scenario-based automated risk assessment device can perform Figure 1 The implementation principle and technical effect of the scenario-based automated risk assessment method described in the illustrated embodiment will not be described in detail. The specific manner in which each module and unit performs operations in the scenario-based automated risk assessment device in the above embodiment has been described in detail in the embodiment of the method, and will not be elaborated here.

[0278] In one possible design, Figure 2 The scenario-based automated risk assessment device of the embodiment shown can be implemented as a computing device, such as Figure 3 As shown, the computing device may include a storage component 31 and a processing component 32;

[0279] The storage component 31 stores one or more computer instructions, wherein the one or more computer instructions are called and executed by the processing component 32 .

[0280] The processing component 32 is used to: obtain input data of the scenario to be evaluated; select at least one risk scenario that matches the input data from a preset risk scenario library; determine the risk factor of the scenario to be evaluated based on the at least one risk scenario; use a risk assessment model to calculate the risk level of the scenario to be evaluated based on the risk factor; and output the risk level.

[0281] The processing component 32 may include one or more processors to execute computer instructions to complete all or part of the steps in the above method. Of course, the processing component may also be implemented by one or more application-specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field programmable gate arrays (FPGAs), controllers, microcontrollers, microprocessors or other electronic components to perform the above method.

[0282] The storage component 31 is configured to store various types of data to support operations at the terminal. The storage component can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, magnetic disk or optical disk.

[0283] Of course, the computing device may also include other components, such as input / output interfaces, display components, communication components, etc.

[0284] The input / output interface provides an interface between the processing component and the peripheral interface module, which may be an output device, an input device, etc.

[0285] The communication component is configured to facilitate, among other things, wired or wireless communications between the computing device and other devices.

[0286] Among them, the computing device can be a physical device or an elastic computing host provided by a cloud computing platform, etc. In this case, the computing device can refer to a cloud server, and the above-mentioned processing components, storage components, etc. can be basic server resources rented or purchased from the cloud computing platform.

[0287] The present application also provides a computer storage medium storing a computer program, wherein the computer program can achieve the above-mentioned Figure 1 The illustrated embodiment is a scenario-based automated risk assessment method.

[0288] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0289] The device embodiments described above are merely illustrative, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. Ordinary technicians in this field can understand and implement it without paying creative labor.

[0290] Through the description of the above implementation methods, those skilled in the art can clearly understand that each implementation method can be implemented by means of software plus a necessary general hardware platform, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solution is essentially or the part that contributes to the prior art can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a disk, an optical disk, etc., including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0291] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit it. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A scenario-based automated risk assessment method. It is characterized in that include: Obtain input data for the scenario to be evaluated; Selecting at least one risk scenario matching the input data from a preset risk scenario library; Determining, according to the at least one risk scenario, a risk factor of the scenario to be assessed; Calculating the risk level of the scenario to be assessed based on the risk factors using a risk assessment model; The risk level is output.

2. The method according to claim 1, It is characterized in that The selecting at least one risk scenario matching the input data from a preset risk scenario library comprises: Performing feature extraction on the input data to obtain a feature vector; In the risk scenario library, at least one risk scenario whose similarity to the feature vector satisfies a similarity condition is searched.

3. The method according to claim 2, It is characterized in that The step of searching for at least one risk scenario whose similarity to the feature vector satisfies a similarity condition comprises: Calculate the cosine similarity between the feature vector and the feature vector of each risk scenario, and select the risk scenario whose cosine similarity meets the similarity condition; or, Calculating the cosine similarity between the feature vector and the feature vector of each risk scenario; Calculating the Manhattan distance between the feature vector and the feature vector of the selected risk scenario, and calculating the degree of matching between the feature vector and the feature vector of the selected risk scenario based on the Manhattan distance; Among them, through the formula: Calculate the matching degree between the feature vector and the feature vector of the selected risk scenario, S represents the matching degree between the feature vector and the feature vector of the selected risk scenario, α is an adjustment parameter used to balance the influence of cosine similarity and Manhattan distance; D represents Manhattan distance; Using the weighted average of the cosine similarity and the matching degree as the comprehensive similarity, and screening out risk scenarios whose comprehensive similarities meet similarity conditions; The comprehensive similarity is calculated by the formula: CS=β·C+(1-β)·S, where CS represents the comprehensive similarity, β is a preset weighting coefficient, C represents the cosine similarity, and S represents the matching degree.

4. The method according to claim 1, It is characterized in that The determining, according to the at least one risk scenario, a risk factor of the scenario to be evaluated comprises: identifying key variables in the at least one risk scenario; Mapping the key variables to the scenario to be evaluated to obtain risk factors; According to the distribution characteristics of risk factors, Bayesian estimation is used to update the probability distribution of risk factors.

5. The method according to claim 1, It is characterized in that The step of calculating the risk level of the scenario to be evaluated according to the risk factor using the risk evaluation model includes: The risk factor is input into a pre-trained risk assessment model, and the risk level is output using the risk assessment model.

6. The method according to claim 5, It is characterized in that Outputting the risk level using the risk assessment model includes: The risk level is calculated using the following formula: Among them, R represents the risk level, f i represents the i-th risk factor, w i is with f i The corresponding weight coefficient, μ i and σ i represent the expected value and standard deviation of the ith risk factor respectively.

7. The method according to claim 6, It is characterized in that The weight coefficient w i Determined by the following formula: Among them, a i is the preset adjustment coefficient, m i represents the ideal value and allowable range of the ith risk factor, r i represents the permissible range of the ith risk factor, i.e., f i With m i The allowable deviation between .

8. The method according to claim 1, It is characterized in that The output of the risk level includes: The risk level is displayed on a user interface; or the risk level is sent to a remote server, so that the remote server performs corresponding processing based on the risk level.

9. A computing device, It is characterized in that It includes a processing component and a storage component; the storage component stores one or more computer instructions; the one or more computer instructions are used to be called and executed by the processing component to implement the scenario-based automated risk assessment method as described in any one of claims 1 to 7.

10. A computer storage medium, It is characterized in that A computer program is stored, and when the computer program is executed by a computer, the scenario-based automated risk assessment method according to any one of claims 1 to 7 is implemented.

Citation Information

Patent Citations

  • Risk assessment report generation method and device, equipment and storage medium

    CN110619568A

  • Information processing method and device and storage medium

    CN113674010A

  • Urban motor vehicle exhaust excessive emission risk prediction and quantitative evaluation method

    CN114548759A

  • Multi-risk analysis and evaluation method and system for water resource coupling system

    CN116090839A

  • Three-dimensional space safety risk assessment method

    CN116579598A

Cited By

  • Aquatic animal invasion risk assessment system, method and device and electronic equipment

    CN120355227A