Fund investment risk assessment user interaction method and system

Abstract

The invention relates to the technical field of financial science and technology, and discloses a fund investment risk assessment user interaction method and system, and the method comprises the steps: collecting the business domain flow state data and historical retracement features of a target fund of a user, and generating a multi-dimensional risk assessment framework of the user in investment; constructing an income-risk association map of the target fund, and constructing a dynamic risk conduction system of the user combination; calculating a decision deviation degree of the user in the historical transaction record, and generating a fund risk coupling index of the user; scheduling a strategy change trajectory and quotient domain tail event data in a historical risk assessment record of the user, quantifying a semantic transition degree of risk preference of the user, and simulating an asset risk resonance effect of an investment portfolio of the user; and performing optimization processing on the multi-dimensional risk evaluation framework to obtain a target evaluation framework, generating a risk hotspot distribution diagram of the user, and generating a dynamic risk topology report of the user. According to the invention, the fund investment risk assessment accuracy can be improved.

Description

Technical Field

[0001] The present invention relates to a user interaction method and system for fund investment risk assessment, belonging to the technical field of financial technology. Background Art

[0002] User-personalized financial advisory interaction refers to a highly customized and interactive communication method between financial institutions or related systems and users in financial advisory services. In this interactive mode, users can elaborate on their unique financial situation in detail, including asset size, income stability, debt situation, etc., which facilitates subsequent fund investment risk assessment and processing.

[0003] The existing user personalized financial consulting interaction methods mainly adopt the human-led interaction method and the questionnaire-algorithm combination method. The human-led interaction method mainly involves users making appointments with financial advisors, communicating their own financial status, investment goals and other information face-to-face or online, and advisors analyzing and providing users with personalized solutions based on their professional knowledge and experience. The disadvantage of this method is that it is limited by the advisor's personal energy, the service efficiency is low, and due to personal cognitive differences, the accuracy and consistency of evaluations and suggestions are difficult to guarantee, and the consulting cost is high. The questionnaire-algorithm combination method is that the user first fills out a detailed questionnaire covering various information such as finance and risk preferences. The system analyzes the questionnaire data through an algorithm and generates a preliminary consulting report. However, this method relies too much on the user's understanding and accuracy of the questionnaire, and the algorithm is relatively fixed, and cannot flexibly respond to the user's complex and changing needs in real time. The questionnaire content may be lengthy, affecting the user experience, and the generated report lacks personal depth and adjustment. Therefore, a method is needed to improve the accuracy of fund investment risk assessment. Summary of the Invention

[0004] The present invention provides a fund investment risk assessment user interaction method and system, the main purpose of which is to improve the accuracy of fund investment risk assessment.

[0005] To achieve the above objectives, the present invention provides a user interaction method for fund investment risk assessment, comprising: Obtaining the user's investment interaction configuration plan and risk tolerance parameters when interacting with the fund investment system, collecting the business domain flow data and historical drawdown characteristics of the user's target fund, and generating a multi-dimensional risk assessment framework for the user's investment based on the risk tolerance parameters; Constructing a return-risk correlation map of the target fund based on the interactive investment configuration plan and the historical drawdown characteristics, and constructing a dynamic risk transmission system for the user portfolio based on the business domain flow data; Calculating the decision deviation of the user in historical transaction records, and calculating the probability of extreme value exposure of the target fund under different market pressure scenarios, and combining the decision deviation and the probability of extreme value exposure to generate the user's fund risk coupling index; Dispatching the strategy change trajectory and business domain tail event data in the user's historical risk assessment records, quantifying the semantic change degree of the user's risk preference based on the strategy change trajectory, and simulating the asset risk resonance effect of the user's investment portfolio based on the business domain tail event data; Combined with the fund risk coupling index, the semantic change degree and the asset risk resonance effect, the multidimensional risk assessment framework is optimized to obtain a target assessment framework. Based on the dynamic risk transmission system, a risk hotspot distribution map of the user is generated. Based on the risk hotspot distribution map and the target assessment framework, a dynamic risk topology report of the user is generated.

[0006] Optionally, generating a multi-dimensional risk assessment framework for the user's investment based on the risk tolerance parameter includes: Performing interval analysis on the risk tolerance parameter to obtain the risk tolerance interval of the user; Discretizing the risk tolerance interval to obtain a risk quantification level; Based on a preset rule engine, multi-dimensional projection processing is performed on the risk quantification level to obtain a multi-dimensional risk matrix; The multi-dimensional risk matrix is ​​subjected to strategy adaptation processing to generate a multi-dimensional risk assessment framework for the user's investment.

[0007] Optionally, constructing a return-risk correlation map of the target fund based on the investment interaction configuration scheme and the historical drawdown characteristics includes: Performing information extraction processing on the input interaction configuration scheme to obtain key interaction elements; Performing modal decomposition on the historical drawdown characteristics to obtain modal components of risk factors; Performing a benefit causal relationship analysis on the risk factor modal components to obtain a risk-benefit causal link; Identifying risk-benefit factor nodes and directed causal relationship edges in the risk-benefit causal link; Based on the risk-return factor nodes and the causal relationship directed edges, a return-risk association graph of the target fund is constructed.

[0008] Optionally, constructing a dynamic risk transmission system for the user portfolio based on the business domain flow data includes: Extracting the business domain multidimensional features corresponding to the business domain flow state data, screening out business domain risk features from the business domain multidimensional features, and calculating the time feature correlation between the business domain risk features; Quantifying the business domain risk feature to obtain a risk feature value, and calculating a time feature attenuation rate corresponding to the business domain risk feature based on the risk feature value; Calculate the risk contagion intensity corresponding to the business domain risk feature by combining the time feature correlation and the feature decay rate; Identify investment portfolio elements of the user portfolio, and construct a dynamic risk transmission system for the user portfolio based on the risk contagion intensity and the investment portfolio elements.

[0009] Optionally, the calculating the risk contagion intensity corresponding to the business domain risk feature by combining the time feature correlation degree and the time feature decay rate includes: ; in, represents the risk contagion intensity corresponding to the business domain risk characteristics, t represents the attention time length of the business domain risk characteristics, Indicates the starting time of attention to business domain risk characteristics, It represents the time feature correlation between the i-th feature and the j-th feature in the business domain risk feature at the starting attention time. It represents the time feature attenuation rate of the i-th feature and the j-th feature in the business domain risk feature at the starting attention time. i and j represent the serial numbers corresponding to the business domain risk features, and i and j have different values. n represents the number of business domain risk features.

[0010] Optionally, calculating the decision deviation of the user in historical transaction records includes: Collecting initial transaction data of the user in historical transaction records; Performing data cleaning on the initial transaction data to obtain target transaction data; Counting the indicator transaction features corresponding to the target transaction data, and querying the preset transaction criteria corresponding to each indicator in the indicator transaction features; Based on the preset transaction criteria, constructing a transaction reference vector corresponding to the user; The decision deviation of the user in the historical transaction records is calculated by combining the transaction benchmark vector and the indicator transaction feature.

[0011] Optionally, the calculating the decision deviation of the user in the historical transaction records by combining the transaction benchmark vector and the indicator transaction feature includes: Performing feature coding processing on the indicator transaction feature to obtain a feature coding value; Normalizing the transaction benchmark vector to obtain a standard benchmark vector, and calculating a benchmark volatility corresponding to the standard benchmark vector; Combining the standard reference vector, the feature code value, and the reference volatility, the decision deviation of the user in the historical transaction record can be calculated using the following formula: ; Among them, A represents the decision deviation of the user in the historical transaction records, Indicates the characteristic code value of the ath indicator in the indicator trading characteristics, Represents the standard basis vector of the ath indicator in the indicator trading characteristics, represents the benchmark volatility of the standard benchmark vector of the a-th indicator, a represents the indicator serial number corresponding to the indicator trading feature, and q represents the number of indicators.

[0012] Optionally, quantifying the semantic change degree of the user's risk preference based on the strategy change trajectory includes: Extracting policy metadata from the policy change trajectory, and calculating data entropy corresponding to the policy metadata; identifying key policy information from the policy metadata based on the data entropy; Performing vectorization processing on the key strategy information to obtain a strategy information vector; The vector differences between adjacent vectors in the strategy information vector are calculated, and the semantic change degree of the user's risk preference is quantified based on the vector differences.

[0013] Optionally, simulating the asset risk resonance effect of the user's investment portfolio based on the business domain tail event data includes: Performing data enhancement processing on the business domain tail event data to obtain target tail event data; Analyzing the event identifier corresponding to the target tail event data, and querying the impact event identification rules from the user's capital investment system; identifying a tail impact event in the target tail event data based on the event identifier and the impact event identification rule; calculating an event correlation coefficient between the tail impact events, and determining a risk resonance intensity between the tail impact events based on the event correlation coefficient; Based on the tail shock event and the risk resonance intensity, an asset risk resonance effect of the user's investment portfolio is simulated.

[0014] In order to solve the above problems, the present invention further provides a fund investment risk assessment user interaction system, the system comprising: A multi-dimensional risk assessment framework generation module is used to obtain the user's investment interaction configuration plan and risk tolerance parameters when interacting with the fund investment system, collect the business domain flow data and historical drawdown characteristics of the user's target fund, and generate a multi-dimensional risk assessment framework for the user's investment based on the risk tolerance parameters; A dynamic risk transmission system construction module is used to construct a return-risk correlation map of the target fund based on the investment interaction configuration plan and the historical drawdown characteristics, and to construct a dynamic risk transmission system for the user portfolio based on the business domain flow data; a fund risk coupling index generation module, configured to calculate the decision deviation of the user in historical transaction records, calculate the probability of extreme value exposure of the target fund under different market pressure scenarios, and generate the user's fund risk coupling index by combining the decision deviation and the extreme value exposure probability; An asset risk resonance effect simulation module is used to schedule the strategy change trajectory and business domain tail event data in the user's historical risk assessment records, quantify the semantic change degree of the user's risk preference based on the strategy change trajectory, and simulate the asset risk resonance effect of the user's investment portfolio based on the business domain tail event data; A risk report generation module is used to optimize the multidimensional risk assessment framework by combining the fund risk coupling index, the semantic change degree and the asset risk resonance effect to obtain a target assessment framework, generate a risk hotspot distribution map of the user based on the dynamic risk transmission system, and generate a dynamic risk topology report of the user based on the risk hotspot distribution map and the target assessment framework.

[0015] Compared with the problems described in the background technology, the present invention can grasp the user's personalized investment needs and risk tolerance boundaries by obtaining the investment interaction configuration plan and risk tolerance parameters of the user when interacting with the investment in the fund investment system, and collect the business domain flow data and historical drawdown characteristics of the user's target fund, so as to understand the market performance and potential risk status of the target fund, and provide a basis for subsequent related analysis. Furthermore, the present invention constructs the return-risk correlation map of the target fund according to the investment interaction configuration plan and the historical drawdown characteristics, which can help the user to intuitively compare the return prospects and risk status of the target fund under different investment arrangements, so that the user can clearly grasp the relationship between key investment variables and thus optimize the investment strategy. The present invention calculates the decision deviation of the user in the historical transaction records, which can be obtained through the The decision deviation degree understands the degree of difference between the user's past transaction decisions and the general market rules or expectations, assists in discovering irrational behaviors, potential risk points and investment style characteristics in transactions, and provides a reference for optimizing investment decisions and risk management. Furthermore, the present invention quantifies the semantic change degree of the user's risk preference based on the strategy change trajectory, and can understand the dynamic change trend of the user's risk preference over time, and assist financial institutions in providing users with personalized financial products and services that match their latest risk tolerance level and investment tendencies. Furthermore, the present invention optimizes the multidimensional risk assessment framework by combining the fund risk coupling index, the semantic change degree and the asset risk resonance effect, thereby improving the accuracy of the multidimensional risk assessment framework and improving the accuracy of the subsequent dynamic risk topology report of the user. Therefore, the fund investment risk assessment user interaction method and system provided in the embodiment of the present invention can improve the accuracy of fund investment risk assessment. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 A schematic diagram of a flow chart of a user interaction method for fund investment risk assessment provided by one embodiment of the present invention; Figure 2 A schematic diagram of modules for implementing the user interaction method for fund investment risk assessment provided by one embodiment of the present invention.

[0017] The purpose, features and advantages of the present invention will be further described with reference to the accompanying drawings and in conjunction with the embodiments. DETAILED DESCRIPTION

[0018] It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0019] The present embodiment provides a user interaction method for fund investment risk assessment. The execution entity of this method includes, but is not limited to, at least one of a server, a terminal, or other electronic device capable of executing the method provided in the present embodiment. In other words, the method can be executed by software or hardware installed on a terminal device or a server device. The server may include, but is not limited to, a single server, a server cluster, a cloud server, or a cloud server cluster.

[0020] Example 1: Reference Figure 1 FIG. 1 is a flow chart of a user interaction method for fund investment risk assessment according to an embodiment of the present invention. In this embodiment, the user interaction method for fund investment risk assessment includes: S1. Obtain the user's investment interaction configuration plan and risk tolerance parameters when interacting with the fund investment system, collect the business domain flow data and historical drawdown characteristics of the user's target fund, and generate a multi-dimensional risk assessment framework for the user's investment based on the risk tolerance parameters.

[0021] The present invention can grasp the user's personalized investment needs and risk tolerance boundaries by obtaining the investment interaction configuration plan and risk tolerance parameters when the user interacts with the investment in the fund investment system, and collect the business domain flow data and historical drawdown characteristics of the user's target fund to understand the market performance and potential risk status of the target fund, providing a basis for subsequent related analysis, wherein the investment interaction configuration plan is the fund allocation plan when the user interacts with the investment in the fund investment system, covering the proportion of the amount invested in different fund products, investment cycle arrangement, etc.; the risk tolerance parameter is the investment risk tolerance and risk tolerance formed by the user based on factors such as his own financial situation, investment experience and psychological expectations. The target fund is a quantitative reflection of the user's intention; the target fund is a specific fund product that the user plans to invest in and expects to obtain returns; the business domain flow data is the real-time dynamic information of the target fund in the commercial activities of the financial market, such as transaction price fluctuations, capital flow trends, etc.; the historical drawdown characteristics are a summary of the performance of the target fund in the past time period, such as the decline in net value, frequency and duration; further, the user's investment interaction configuration plan and risk tolerance parameters can be obtained by the user filling out a questionnaire independently and connecting to the personal financial account data interface; the collection of the business domain flow data and historical drawdown characteristics of the user's target fund can be achieved by connecting to the financial data service platform interface.

[0022] The present invention dynamically generates a multi-dimensional risk assessment framework for user investment based on the risk tolerance parameters, which can provide users with accurate risk assessments that suit their own risk tolerance, help them plan their investments rationally, and effectively avoid potential risks. It should be explained that the multi-dimensional risk assessment framework is a comprehensive assessment system used by users to comprehensively and accurately measure the various risks involved in the target investment funds when investing.

[0023] Specifically, generating a multi-dimensional risk assessment framework for the user's investment based on the risk tolerance parameter includes: Performing interval analysis on the risk tolerance parameter to obtain the risk tolerance interval of the user; Discretizing the risk tolerance interval to obtain a risk quantification level; Based on a preset rule engine, multi-dimensional projection processing is performed on the risk quantification level to obtain a multi-dimensional risk matrix; The multi-dimensional risk matrix is ​​subjected to strategy adaptation processing to generate a multi-dimensional risk assessment framework for the user's investment.

[0024] The risk tolerance interval is the range limit obtained by performing interval analysis on the risk tolerance parameter, representing the range limit of the risk that the user can bear during the investment process. For example, from the perspective of investment loss, the user may be able to accept a loss of -5% to 10% of assets within a certain period of time. This range is their risk tolerance interval. The risk quantification level is the level obtained by discretizing the risk tolerance interval. For example, the risk tolerance interval is divided into five levels: low risk, medium-low risk, medium risk, medium-high risk, and high risk, each level corresponding to a certain range of risk levels. The preset rule engine is a system component that pre-sets a series of rules and logic. The multidimensional risk matrix is ​​a multidimensional matrix structure obtained by multi-dimensional projection processing of the risk quantification level. The risk quantification level is displayed and analyzed from multiple dimensions. For example, the risk situation can be comprehensively considered from the dimensions of investment product type, market environment, investment period, etc., making the risk assessment more comprehensive and three-dimensional.

[0025] Furthermore, the risk tolerance parameter can be analyzed in intervals through statistical analysis and threshold setting algorithms to obtain the risk tolerance interval of the user, such as by analyzing the fluctuation range of the user's past investment behavior data, combining the risk tolerance limit of the user's subjective feedback, using statistical analysis methods to determine the key data characteristic values, and then setting reasonable thresholds based on financial investment theory and industry experience, so as to accurately define the upper and lower limits of the risk tolerance interval; the risk tolerance interval can be discretized through an equidistant partitioning algorithm to obtain a risk quantification level; based on a preset rule engine, the risk quantification level can be multi-dimensionally projected using multi-dimensional mapping rules and association analysis to obtain a multi-dimensional risk matrix; the multi-dimensional risk matrix can be strategy-adapted by combining the user's investment configuration information with a risk response strategy library, thereby generating a multi-dimensional risk assessment framework for the user's investment.

[0026] S2. Construct a return-risk correlation map of the target fund based on the investment interaction configuration plan and the historical drawdown characteristics, and construct a dynamic risk transmission system of the user portfolio based on the business domain flow data.

[0027] The present invention constructs a return-risk correlation map of the target fund based on the investment interaction configuration plan and the historical drawdown characteristics, which can help users intuitively compare the return prospects and risk status of the target fund under different investment arrangements, allowing users to clearly grasp the relationship between key investment variables and thus optimize investment strategies. Among them, the return-risk correlation map is a nonlinear relationship network between the return and risk factors of the target fund, which is used to quantify the impact intensity and transmission path of different risk factors on the fund return.

[0028] In detail, the constructing of the return-risk correlation map of the target fund based on the investment interaction configuration scheme and the historical drawdown characteristics includes: Performing information extraction processing on the input interaction configuration scheme to obtain key interaction elements; Performing modal decomposition on the historical drawdown characteristics to obtain modal components of risk factors; Performing a benefit causal relationship analysis on the risk factor modal components to obtain a risk-benefit causal link; Identifying risk-benefit factor nodes and directed causal relationship edges in the risk-benefit causal link; Based on the risk-return factor nodes and the causal relationship directed edges, a return-risk association graph of the target fund is constructed.

[0029] Among them, the key interactive factors are the core investment components identified by information extraction of the investment interaction configuration plan; the risk factor modal components are independent components of different risk types separated after modal decomposition of the historical drawdown characteristics; the risk-return causal link is the path context of risk affecting return sorted out after the risk factor modal components are subjected to return causal relationship analysis; the risk-return factor nodes represent the key elements such as risk factors and returns in the risk-return causal link, and the directed edges of the causal relationship reflect the direction of the causal effect between the elements.

[0030] Furthermore, the input interaction configuration scheme can be subjected to information extraction processing by an OCR recognition method to obtain key interaction elements; The historical drawdown characteristics can be modally decomposed using a variational mode decomposition (VMD) algorithm to obtain risk factor modal components; the risk factor modal components can be subjected to a return causal relationship analysis using statistical analysis methods such as the Granger causality test to obtain a risk-return causal link; the risk-return factor nodes and causal directed edges in the risk-return causal link can be identified using a node and edge identification algorithm; based on the risk-return factor nodes and the causal directed edges, a return-risk association graph of the target fund can be constructed using graph construction technology, such as a graph generation technology based on a network analysis library (such as NetworkX). With the help of this technology, the risk-return factor nodes are defined as node objects in the graph, and the causal directed edges are set as directed edges connecting the nodes. Based on the risk and return related attribute information carried by the nodes and edges, such as the degree of influence of the risk factors, the fluctuation range of the returns, etc., the layout of the graph is optimized and the visualization is set, thereby clearly presenting the complex return-risk association relationship of the target fund.

[0031] The present invention constructs a dynamic risk transmission system for the user portfolio based on the business domain flow data, which can monitor the impact of subtle market changes on the user's investment portfolio in real time, accurately locate the risk source and transmission path, and warn the user of potential risks in advance. Among them, the dynamic risk transmission system is a real-time risk tracking and early warning mechanism for the user portfolio.

[0032] Specifically, the dynamic risk transmission system of the user portfolio is constructed based on the business domain flow data, including: Extracting the business domain multidimensional features corresponding to the business domain flow state data, screening out business domain risk features from the business domain multidimensional features, and calculating the time feature correlation between the business domain risk features; Quantifying the business domain risk feature to obtain a risk feature value, and calculating a time feature attenuation rate corresponding to the business domain risk feature based on the risk feature value; Calculate the risk contagion intensity corresponding to the business domain risk feature by combining the time feature correlation and the feature decay rate; Identify investment portfolio elements of the user portfolio, and construct a dynamic risk transmission system for the user portfolio based on the risk contagion intensity and the investment portfolio elements.

[0033] Among them, the business domain multidimensional features are feature information of multiple dimensions corresponding to the business domain flow data, which comprehensively reflect different aspects of the business domain flow data; the business domain risk features are risk-related features screened out from the business domain multidimensional features, which can reflect the potential risk status in the business domain; the time feature correlation indicates the degree of temporal correlation between the business domain risk features, reflecting the dependence or influence relationship between the risk features; the risk feature value is the numerical value obtained after quantifying the business domain risk features, so as to facilitate numerical analysis and comparison of risk features; the time feature decay rate indicates the speed at which the influence corresponding to the business domain risk features weakens over time, and is used to measure the timeliness of the risk features; the risk contagion intensity indicates the intensity of the risk corresponding to the business domain risk features spreading and influencing between different features, and comprehensively considers the feature correlation and feature decay rate; the investment portfolio elements are specific components of the user portfolio, which are various types of assets or investment targets invested by the user.

[0034] Furthermore, data mining algorithms (such as association rule mining and principal component analysis) can be used to extract multidimensional business domain features corresponding to the business domain flow data. Business domain risk features can be screened out from the multidimensional business domain features using domain knowledge and machine learning classification algorithms (such as logistic regression and support vector machines). The temporal characteristic correlation between the business domain risk features can be calculated using correlation analysis methods (such as the Pearson correlation coefficient and the Spearman correlation coefficient). The business domain risk features can be quantified using data preprocessing techniques such as normalization and standardization to obtain risk characteristic values. Based on the risk characteristic values, a time series analysis algorithm (such as exponential smoothing and an autoregressive moving average model) can be used to calculate the temporal characteristic decay rate corresponding to the business domain risk features. The investment portfolio elements of the user portfolio can be identified by combing through the investment portfolio list or querying the user's investment records. Based on the risk contagion intensity and the investment portfolio elements, a dynamic risk transmission system for the user portfolio can be constructed using graph theory modeling techniques (such as constructing a directed weighted graph with the investment portfolio elements as nodes and the risk contagion intensity as the edge weights).

[0035] Furthermore, as an optional embodiment of the present invention, the step of calculating the risk contagion intensity corresponding to the business domain risk feature by combining the time feature correlation degree and the time feature decay rate includes: ; in, represents the risk contagion intensity corresponding to the business domain risk characteristics, t represents the attention time length of the business domain risk characteristics, Indicates the starting time of attention to business domain risk characteristics, It represents the time feature correlation between the i-th feature and the j-th feature in the business domain risk feature at the starting attention time. It represents the time feature attenuation rate of the i-th feature and the j-th feature in the business domain risk feature at the starting attention time. i and j represent the serial numbers corresponding to the business domain risk features, and i and j have different values. n represents the number of business domain risk features.

[0036] The above formula quantifies the intensity of risk transmission between business domain risk characteristics by comprehensively considering the correlation of time characteristics, the decay rate of time characteristics, and the weights of different time points.

[0037] S3. Calculate the decision deviation of the user in historical transaction records, and calculate the probability of extreme value exposure of the target fund under different market pressure scenarios. Combine the decision deviation and the probability of extreme value exposure to generate a fund risk coupling index for the user.

[0038] By calculating the decision deviation of the user in the historical transaction records, the present invention can understand the degree of difference between the user's past transaction decisions and the general market rules or expectations through the decision deviation, assist in discovering irrational behaviors, potential risk points and investment style characteristics in transactions, and provide a reference for optimizing investment decisions and risk management, wherein the decision deviation indicates the degree of difference between the actual decision of the user in the historical transaction records and reference standards such as market norms, expected returns or reasonable investment strategies.

[0039] Specifically, calculating the decision deviation of the user in the historical transaction records includes: Collecting initial transaction data of the user in historical transaction records; Performing data cleaning on the initial transaction data to obtain target transaction data; Counting the indicator transaction features corresponding to the target transaction data, and querying the preset transaction criteria corresponding to each indicator in the indicator transaction features; Based on the preset transaction criteria, constructing a transaction reference vector corresponding to the user; The decision deviation of the user in the historical transaction records is calculated by combining the transaction benchmark vector and the indicator transaction feature.

[0040] Among them, the initial transaction data is a collection of all the original transaction information of the user in the historical transaction records without any processing, covering various basic data such as the time, type, quantity, price, etc. of the transaction, for example, the specific time, name, quantity and transaction price of each stock transaction of the user in the stock market in the past year; the target transaction data is the high-quality transaction data after the initial transaction data has been cleaned, duplicate values ​​have been removed, erroneous data have been corrected, and missing values ​​have been filled or eliminated; the indicator transaction characteristics are the quantitative data characteristics corresponding to the target transaction data that can reflect the key characteristics of the transaction and are used for analysis and evaluation, such as the average profit amount per transaction, transaction frequency, and duration extracted from the target transaction data. The preset trading criteria are a series of rules corresponding to each indicator in the indicator trading characteristics, which are predetermined and used to guide one's own trading behavior and judge the rationality of the transaction. For example, the user sets the rule of only investing in stocks with a price-to-earnings ratio of less than 20 times, and the maximum loss of a single transaction cannot exceed 5% of the invested funds. The trading benchmark vector is a numerical vector form that converts the preset trading criteria into a quantitative comparison, representing a standard reference that meets the user's expected trading behavior. For example, if the user's preset trading criteria are that the annualized return on investment in stocks must reach 15% and the stock turnover rate does not exceed 300%, then the trading benchmark vector can be expressed as [annualized return 15%, turnover rate 300%] for comparison and analysis with actual trading data.

[0041] Optionally, the initial transaction data of the user in the historical transaction records can be collected from a third-party payment platform, such as Alipay or WeChat Pay; the initial transaction data can be cleaned by the box plot method to obtain the target transaction data; the indicator transaction characteristics corresponding to the target transaction data can be counted by basic statistical calculation methods, such as the mean, median and standard deviation; the preset transaction criteria corresponding to the user can be queried through the user's personal investment file; based on the preset transaction criteria, the direct numerical mapping method is used to construct the transaction benchmark vector corresponding to the user. If the preset transaction criteria are clear numerical indicators, these numerical values ​​can be directly arranged in a certain order to form a vector. For example, the preset transaction criteria are an annualized rate of return of not less than 10% and a maximum drawdown of not more than 20%.

[0042] Furthermore, as an optional embodiment of the present invention, the combining of the transaction benchmark vector and the indicator transaction feature to calculate the decision deviation of the user in the historical transaction record includes: Performing feature coding processing on the indicator transaction feature to obtain a feature coding value; Normalizing the transaction benchmark vector to obtain a standard benchmark vector, and calculating a benchmark volatility corresponding to the standard benchmark vector; Combining the standard reference vector, the feature code value, and the reference volatility, the decision deviation of the user in the historical transaction record can be calculated using the following formula: ; Among them, A represents the decision deviation of the user in the historical transaction records, Indicates the characteristic code value of the ath indicator in the indicator trading characteristics, Represents the standard basis vector of the ath indicator in the indicator trading characteristics, represents the benchmark volatility of the standard benchmark vector of the a-th indicator, a represents the indicator serial number corresponding to the indicator trading feature, and q represents the number of indicators.

[0043] Among them, the characteristic coding value is the value of the indicator trading feature encoded in a specific way, the standard benchmark vector is the vector of the trading benchmark vector after standardization, and the benchmark volatility is the specific numerical value of the volatility indicator corresponding to the standard benchmark vector.

[0044] Furthermore, the indicator transaction characteristics can be feature-encoded using an encoding algorithm to obtain a feature encoding value, such as a one-hot encoding algorithm; the transaction benchmark vector can be standardized using a maximum-minimum normalization method to obtain a standard benchmark vector; and the benchmark volatility corresponding to the standard benchmark vector can be calculated by applying statistical methods such as standard deviation to the fluctuation of data at each time point in the standard benchmark vector.

[0045] The present invention generates the user's fund risk coupling index by combining the decision deviation and the extreme value exposure probability, which can accurately evaluate the comprehensive risk status faced by the user in fund investment, provide investors with a more comprehensive and detailed risk reference, and help them optimize investment decisions and reduce potential losses. Among them, the extreme value exposure probability is the probability of the target fund experiencing extreme returns under different market pressure scenarios, such as the possibility of large fluctuations in the fund's net value under extreme circumstances such as a sharp drop or rise in the market; the fund risk coupling index is a quantitative indicator after comprehensive consideration of the interaction between the user's investment risk and the target fund risk, such as reflecting the user's own The overall risk level is jointly determined by trading decisions and the risk characteristics of funds under different market conditions. Furthermore, the statistics of the extreme value exposure probability of the target fund under different market pressure scenarios can be achieved through historical data back-analysis, simulation of fund net value fluctuations under extreme market conditions and the use of probability statistical models; combining the decision deviation and the extreme value exposure probability to generate the user's fund risk coupling index, such as multiplying the decision deviation by a coefficient reflecting the weight of the influence of the user's trading behavior, and adding it to the standardized extreme value exposure probability to obtain a fund risk coupling index that comprehensively reflects the user's investment behavior and the fund's risk status.

[0046] S4. Dispatching the strategy change trajectory and business domain tail event data in the user's historical risk assessment records, quantifying the semantic change degree of the user's risk preference based on the strategy change trajectory, and simulating the asset risk resonance effect of the user's investment portfolio based on the business domain tail event data.

[0047] The present invention quantifies the semantic change degree of the user's risk preference based on the strategy change trajectory, thereby understanding the dynamic change trend of the user's risk preference over time and assisting financial institutions in providing users with personalized financial products and services that match their latest risk tolerance level and investment tendency. The strategy change trajectory is the clue of investment strategy change reflected by the risk assessment of different periods in the user's historical risk assessment record, such as the evolution process from the early conservative strategy to the mid-term balanced strategy and then to the late aggressive strategy; the business domain tail event data is the information related to the extreme risk events with low probability but great impact that occurred in the business field in the user's historical risk assessment record, such as the record of the user's The semantic change degree represents the degree of change of the user's risk preference over time. For example, if the semantic change degree value is large, it means that the user's risk preference has changed significantly from conservative to aggressive or vice versa, such as gradually changing from investing only in low-risk bonds to investing heavily in high-risk stocks. If the value is small, it indicates that the user's risk preference is relatively stable, such as maintaining a sound asset allocation strategy. Furthermore, the strategy change trajectory in the user's historical risk assessment record and the scheduling of business domain tail event data can be achieved through the construction of an associated database index, the use of data mining algorithms for screening, and the combination of user identity identification for targeted retrieval.

[0048] Specifically, quantifying the semantic change degree of the user's risk preference based on the strategy change trajectory includes: Extracting policy metadata from the policy change trajectory, and calculating data entropy corresponding to the policy metadata; identifying key policy information from the policy metadata based on the data entropy; Performing vectorization processing on the key strategy information to obtain a strategy information vector; The vector differences between adjacent vectors in the strategy information vector are calculated, and the semantic change degree of the user's risk preference is quantified based on the vector differences.

[0049] Among them, the strategy metadata is the strategy description text information in the strategy change trajectory, covering asset allocation, investment objectives and other contents, which is used to accurately define the investment strategy; the data entropy represents the degree of information confusion or uncertainty corresponding to the strategy metadata. The higher the data entropy, the greater the uncertainty of the information contained in the strategy metadata; the key strategy information is the information in the strategy metadata such as asset allocation ratio and expected return range that plays a key role in analyzing changes in risk preferences; the strategy information vector is the vector form formed by converting the key strategy information according to specific rules, which is convenient for quantitative analysis; the vector difference represents the difference value between adjacent vectors in the strategy information vector, which is used to measure the degree of change in the risk preference semantics of strategies in adjacent periods.

[0050] Furthermore, the policy metadata in the policy change trajectory can be extracted through text parsing technology; the data entropy corresponding to the policy metadata can be calculated through the entropy calculation method in information theory; based on the data entropy, key policy information can be identified from the policy metadata using a data mining algorithm; the key policy information can be vectorized through feature engineering technology to obtain a policy information vector; the vector difference between adjacent vectors in the policy information vector can be calculated through algorithms such as Euclidean distance or cosine similarity; based on the vector difference, the difference can be cumulatively analyzed in the time dimension to quantify the semantic change degree of the user's risk preference.

[0051] The present invention simulates the asset risk resonance effect of the user's investment portfolio based on the business domain tail event data, and can discover in advance the possibility and extent of risk shocks to various types of assets in the user's investment portfolio under extreme market conditions, helping investors to adjust asset allocation in advance and reduce the probability of significant losses to the investment portfolio caused by systemic risks. The asset risk resonance effect is a phenomenon of the user's investment portfolio, that is, the phenomenon that multiple assets in the portfolio, due to mutual correlation and influence under specific market conditions, simultaneously experience risk conditions such as large price fluctuations or value declines.

[0052] In detail, simulating the asset risk resonance effect of the user's investment portfolio based on the business domain tail event data includes: Performing data enhancement processing on the business domain tail event data to obtain target tail event data; Analyzing the event identifier corresponding to the target tail event data, and querying the impact event identification rules from the user's capital investment system; identifying a tail impact event in the target tail event data based on the event identifier and the impact event identification rule; calculating an event correlation coefficient between the tail impact events, and determining a risk resonance intensity between the tail impact events based on the event correlation coefficient; Based on the tail shock event and the risk resonance intensity, an asset risk resonance effect of the user's investment portfolio is simulated.

[0053] Among them, the target tail event data is event data containing more details and representativeness obtained after data enhancement processing of the business domain tail event data, such as expanding the original data through time series interpolation and similar event feature migration; the event identifier is a coding combination corresponding to the target tail event data for describing event characteristics, such as a timestamp, event category code, etc.; the impact event identification rule is a standard for determining whether an event is a shock event based on the user's investment situation and risk preference queried in the user's capital investment system, for example, for a certain conservative user, a bond yield drop of more than 5 basis points within a week and an investment ratio of more than 30% is considered a shock event; the tail shock event is an event in the target tail event data that meets the shock event identification rule and has a significant impact on the investment portfolio, such as an event that meets the changes in specific economic indicators during a macroeconomic crisis; the risk resonance intensity is an indicator that comprehensively measures the possibility and intensity of risk resonance of investment portfolio assets caused by the tail shock events, for example, calculated by combining the event correlation coefficient with the asset category weight. A high coefficient means a high possibility and intensity of risk resonance.

[0054] Furthermore, the business domain tail event data can be enhanced by methods such as time series interpolation, similar event feature migration, and generative adversarial networks (GANs) to obtain target tail event data; the event identifiers corresponding to the target tail event data can be analyzed by structural decomposition and feature extraction of event identifiers through professional data analysis tools, and the data analysis tools are compiled in a scripting language; the shock event identification rules can be queried from the user's capital investment system through the user's investment account information, past transaction records, and risk preference parameters; based on the event identifiers and the shock event identification rules, the tail shock events in the target tail event data can be identified, such as a macroeconomic crisis event that meets preset rules such as a quarterly GDP growth rate that falls by more than 3% year-on-year; The Pearson correlation coefficient algorithm calculates the event correlation coefficient between the tail shock events, and directly uses the event correlation coefficient as the risk resonance intensity between the tail shock events; based on the tail shock events and the risk resonance intensity, simulates the asset risk resonance effect of the user's investment portfolio; the specific steps are: associating the tail shock events and their risk resonance intensity with the investment portfolio assets, and using Monte Carlo simulation combined with a complex network analysis model to simulate the dynamic changes of asset risk indicators and the spread of risk resonance, such as simulating the yield fluctuations, price fluctuations, and risk transmission paths and scope of various assets such as stocks, bonds, and funds in the investment portfolio when tail shock events occur in different industries with different risk resonance intensity combinations, and then evaluating the changing trend of the overall value of the investment portfolio.

[0055] S5. Combine the fund risk coupling index, the semantic change degree and the asset risk resonance effect to optimize the multidimensional risk assessment framework to obtain a target assessment framework. Based on the dynamic risk transmission system, generate a risk hotspot distribution map of the user. Based on the risk hotspot distribution map and the target assessment framework, generate a dynamic risk topology report of the user.

[0056] The present invention optimizes the multidimensional risk assessment framework by combining the fund risk coupling index, the semantic variability and the asset risk resonance effect, thereby improving the accuracy of the multidimensional risk assessment framework and the accuracy of the subsequent dynamic risk topology report of the user. The target assessment framework is the framework obtained by optimizing the multidimensional risk assessment framework in combination with the fund risk coupling index, the semantic variability and the asset risk resonance effect. Further, the multidimensional risk assessment framework is optimized in combination with the fund risk coupling index, the semantic variability and the asset risk resonance effect to obtain a target assessment framework. First, the fund risk coupling index is incorporated into the framework to quantify the degree of interaction between risks between funds; then the semantic variability is incorporated to analyze the impact of changes in investors' risk preferences on investment decisions; finally, combined with the asset risk resonance effect, the concentrated outbreak of asset risks in extreme markets is considered, the weights of various indicators are comprehensively adjusted, and the target assessment framework is improved.

[0057] The present invention generates a risk hotspot distribution map of the user based on the dynamic risk transmission system, which can intuitively display the areas and key nodes where risks in the user's investment are concentrated, helping investors to quickly locate high-risk sectors. Based on the risk hotspot distribution map and the target evaluation framework, a dynamic risk topology report of the user is generated, and a report that comprehensively and intuitively presents the overall picture of the user's investment risk, the risk transmission path and the comprehensive evaluation results can be obtained, providing investors with a basis for accurate risk decision-making. Among them, the risk hotspot distribution map is an intuitive visual presentation of the distribution status of the user's investment risk in various asset categories, industries or investment fields, and highlights the risk concentration areas through elements such as color, shape, and size. The dynamic risk topology report is a report of the user covering the overall picture of investment risk, risk transmission path, correlation between various risk factors and comprehensive evaluation results. Furthermore, based on the dynamic risk transmission system, the risk hotspot distribution map of the user can be generated by visualization technology. Based on the risk hotspot distribution map and the target evaluation framework, the dynamic risk topology report of the user can be generated by a report generation tool, and the report generation tool is compiled in JAVA language.

[0058] Compared with the problems described in the background technology, the present invention can grasp the user's personalized investment needs and risk tolerance boundaries by obtaining the investment interaction configuration plan and risk tolerance parameters of the user when interacting with the investment in the fund investment system, and collect the business domain flow data and historical drawdown characteristics of the user's target fund, so as to understand the market performance and potential risk status of the target fund, and provide a basis for subsequent related analysis. Furthermore, the present invention constructs the return-risk correlation map of the target fund according to the investment interaction configuration plan and the historical drawdown characteristics, which can help the user to intuitively compare the return prospects and risk status of the target fund under different investment arrangements, so that the user can clearly grasp the relationship between key investment variables and thus optimize the investment strategy. The present invention calculates the decision deviation of the user in the historical transaction records, which can be obtained through the The decision deviation degree understands the degree of difference between the user's past transaction decisions and the general market rules or expectations, assists in discovering irrational behaviors, potential risk points and investment style characteristics in transactions, and provides a reference for optimizing investment decisions and risk management. Furthermore, the present invention quantifies the semantic change degree of the user's risk preference based on the strategy change trajectory, and can understand the dynamic change trend of the user's risk preference over time, and assist financial institutions in providing users with personalized financial products and services that match their latest risk tolerance level and investment tendencies. Furthermore, the present invention optimizes the multidimensional risk assessment framework by combining the fund risk coupling index, the semantic change degree and the asset risk resonance effect, thereby improving the accuracy of the multidimensional risk assessment framework and improving the accuracy of the subsequent dynamic risk topology report of the user. Therefore, the fund investment risk assessment user interaction method and system provided in the embodiment of the present invention can improve the accuracy of fund investment risk assessment.

[0059] Example 2: like Figure 2 FIG. 1 is a functional module diagram of a user interaction system for fund investment risk assessment according to the present invention.

[0060] The fund investment risk assessment user interaction system 200 described in the present invention can be installed in an electronic device. Depending on the functionality implemented, the fund investment risk assessment user interaction system may include a multi-dimensional risk assessment framework generation module 201, a dynamic risk transmission system construction module 202, a fund risk coupling index generation module 203, an asset risk resonance effect simulation module 204, and a risk report generation module 205. A module, also referred to as a unit, is a series of computer program segments that can be executed by an electronic device processor and perform a fixed function, and is stored in the electronic device's memory.

[0061] In the embodiment of the present invention, the functions of each module / unit are as follows: The multi-dimensional risk assessment framework generation module 201 is used to obtain the user's investment interaction configuration plan and risk tolerance parameters when performing investment interactions in the fund investment system, collect the business domain flow data and historical drawdown characteristics of the user's target fund, and generate the user's multi-dimensional risk assessment framework for investment based on the risk tolerance parameters; The dynamic risk transmission system construction module 202 is used to construct a return-risk correlation map of the target fund based on the investment interaction configuration plan and the historical drawdown characteristics, and to construct a dynamic risk transmission system for the user portfolio based on the business domain flow data; The fund risk coupling index generating module 203 is configured to calculate the decision deviation of the user in historical transaction records, calculate the probability of extreme value exposure of the target fund under different market pressure scenarios, and generate the fund risk coupling index of the user by combining the decision deviation and the extreme value exposure probability; The asset risk resonance effect simulation module 204 is used to schedule the strategy change trajectory and business domain tail event data in the user's historical risk assessment record, quantify the semantic change degree of the user's risk preference based on the strategy change trajectory, and simulate the asset risk resonance effect of the user's investment portfolio based on the business domain tail event data; The risk report generation module 205 is used to optimize the multi-dimensional risk assessment framework in combination with the fund risk coupling index, the semantic change degree and the asset risk resonance effect to obtain a target assessment framework, generate a risk hotspot distribution map of the user based on the dynamic risk transmission system, and generate a dynamic risk topology report of the user based on the risk hotspot distribution map and the target assessment framework.

[0062] In detail, each module in the fund investment risk assessment user interaction system 200 in the embodiment of the present invention adopts the same Figure 1 The same technical means are used as the user interaction method for fund investment risk assessment described in , and can produce the same technical effects, so they will not be repeated here.

[0063] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the present invention can be implemented in other specific forms without departing from the spirit or essential characteristics of the present invention.

[0064] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not limiting. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention.

Claims

1. A user interaction method for fund investment risk assessment, characterized in that: The method comprises: Obtaining the user's investment interaction configuration plan and risk tolerance parameters when interacting with the fund investment system, collecting the business domain flow data and historical drawdown characteristics of the user's target fund, and generating a multi-dimensional risk assessment framework for the user's investment based on the risk tolerance parameters; Constructing a return-risk correlation map of the target fund based on the interactive investment configuration plan and the historical drawdown characteristics, and constructing a dynamic risk transmission system for the user portfolio based on the business domain flow data; Calculating the decision deviation of the user in historical transaction records, and calculating the probability of extreme value exposure of the target fund under different market pressure scenarios, and combining the decision deviation and the probability of extreme value exposure to generate the user's fund risk coupling index; Dispatching the strategy change trajectory and business domain tail event data in the user's historical risk assessment records, quantifying the semantic change degree of the user's risk preference based on the strategy change trajectory, and simulating the asset risk resonance effect of the user's investment portfolio based on the business domain tail event data; Combined with the fund risk coupling index, the semantic change degree and the asset risk resonance effect, the multidimensional risk assessment framework is optimized to obtain a target assessment framework. Based on the dynamic risk transmission system, a risk hotspot distribution map of the user is generated. Based on the risk hotspot distribution map and the target assessment framework, a dynamic risk topology report of the user is generated.

2. The user interaction method for fund investment risk assessment according to claim 1, characterized in that: Generating a multi-dimensional risk assessment framework for the user's investment based on the risk tolerance parameter includes: Performing interval analysis on the risk tolerance parameter to obtain the risk tolerance interval of the user; Discretizing the risk tolerance interval to obtain a risk quantification level; Based on a preset rule engine, multi-dimensional projection processing is performed on the risk quantification level to obtain a multi-dimensional risk matrix; The multi-dimensional risk matrix is ​​subjected to strategy adaptation processing to generate a multi-dimensional risk assessment framework for the user's investment.

3. The user interaction method for fund investment risk assessment according to claim 1, characterized in that: The constructing of a return-risk correlation map of the target fund according to the investment interaction configuration scheme and the historical drawdown characteristics includes: Performing information extraction processing on the input interaction configuration scheme to obtain key interaction elements; Performing modal decomposition on the historical drawdown characteristics to obtain modal components of risk factors; Performing a benefit causal relationship analysis on the risk factor modal components to obtain a risk-benefit causal link; Identifying risk-benefit factor nodes and directed causal relationship edges in the risk-benefit causal link; Based on the risk-return factor nodes and the causal relationship directed edges, a return-risk association graph of the target fund is constructed.

4. The user interaction method for fund investment risk assessment according to claim 1, characterized in that: The dynamic risk transmission system of the user combination is constructed based on the business domain flow data, including: Extracting the business domain multidimensional features corresponding to the business domain flow state data, screening out business domain risk features from the business domain multidimensional features, and calculating the time feature correlation between the business domain risk features; Quantifying the business domain risk feature to obtain a risk feature value, and calculating a time feature attenuation rate corresponding to the business domain risk feature based on the risk feature value; Calculate the risk contagion intensity corresponding to the business domain risk feature by combining the time feature correlation and the feature decay rate; Identify investment portfolio elements of the user portfolio, and construct a dynamic risk transmission system for the user portfolio based on the risk contagion intensity and the investment portfolio elements.

5. The user interaction method for fund investment risk assessment according to claim 4, characterized in that: The calculating the risk contagion intensity corresponding to the business domain risk feature by combining the time feature correlation degree and the time feature attenuation rate includes: ； in, represents the risk contagion intensity corresponding to the business domain risk characteristics, t represents the attention time length of the business domain risk characteristics, Indicates the starting time of attention to business domain risk characteristics, It represents the time feature correlation between the i-th feature and the j-th feature in the business domain risk feature at the starting attention time. It represents the time feature attenuation rate of the i-th feature and the j-th feature in the business domain risk feature at the starting attention time. i and j represent the serial numbers corresponding to the business domain risk features, and i and j have different values. n represents the number of business domain risk features.

6. The user interaction method for fund investment risk assessment according to claim 1, characterized in that: The calculating of the decision deviation of the user in the historical transaction records includes: Collecting initial transaction data of the user in historical transaction records; Performing data cleaning on the initial transaction data to obtain target transaction data; Counting the indicator transaction features corresponding to the target transaction data, and querying the preset transaction criteria corresponding to each indicator in the indicator transaction features; Based on the preset transaction criteria, constructing a transaction reference vector corresponding to the user; The decision deviation of the user in the historical transaction records is calculated by combining the transaction benchmark vector and the indicator transaction feature.

7. The user interaction method for fund investment risk assessment according to claim 6, characterized in that: The calculating the decision deviation of the user in the historical transaction records by combining the transaction benchmark vector and the indicator transaction feature includes: Performing feature coding processing on the indicator transaction feature to obtain a feature coding value; Normalizing the transaction benchmark vector to obtain a standard benchmark vector, and calculating a benchmark volatility corresponding to the standard benchmark vector; Combining the standard reference vector, the feature code value, and the reference volatility, the decision deviation of the user in the historical transaction record can be calculated using the following formula: ； Among them, A represents the decision deviation of the user in the historical transaction records, Indicates the characteristic code value of the ath indicator in the indicator trading characteristics, Represents the standard basis vector of the ath indicator in the indicator trading characteristics, represents the benchmark volatility of the standard benchmark vector of the a-th indicator, a represents the indicator serial number corresponding to the indicator trading feature, and q represents the number of indicators.

8. The user interaction method for fund investment risk assessment according to claim 1, characterized in that: The step of quantifying the semantic change degree of the user's risk preference based on the strategy change trajectory includes: Extracting policy metadata from the policy change trajectory, and calculating data entropy corresponding to the policy metadata; identifying key policy information from the policy metadata based on the data entropy; Performing vectorization processing on the key strategy information to obtain a strategy information vector; The vector differences between adjacent vectors in the strategy information vector are calculated, and the semantic change degree of the user's risk preference is quantified based on the vector differences.

9. The user interaction method for fund investment risk assessment according to claim 1, characterized in that: The simulating the asset risk resonance effect of the user's investment portfolio based on the business domain tail event data includes: Performing data enhancement processing on the business domain tail event data to obtain target tail event data; Analyzing the event identifier corresponding to the target tail event data, and querying the impact event identification rules from the user's capital investment system; identifying a tail impact event in the target tail event data based on the event identifier and the impact event identification rule; calculating an event correlation coefficient between the tail impact events, and determining a risk resonance intensity between the tail impact events based on the event correlation coefficient; Based on the tail shock event and the risk resonance intensity, an asset risk resonance effect of the user's investment portfolio is simulated.

10. A fund investment risk assessment user interaction system, characterized in that: The system comprises: A multi-dimensional risk assessment framework generation module is used to obtain the user's investment interaction configuration plan and risk tolerance parameters when interacting with the fund investment system, collect the business domain flow data and historical drawdown characteristics of the user's target fund, and generate a multi-dimensional risk assessment framework for the user's investment based on the risk tolerance parameters; A dynamic risk transmission system construction module is used to construct a return-risk correlation map of the target fund based on the investment interaction configuration plan and the historical drawdown characteristics, and to construct a dynamic risk transmission system for the user portfolio based on the business domain flow data; a fund risk coupling index generation module, configured to calculate the decision deviation of the user in historical transaction records, calculate the probability of extreme value exposure of the target fund under different market pressure scenarios, and generate the user's fund risk coupling index by combining the decision deviation and the extreme value exposure probability; An asset risk resonance effect simulation module is used to schedule the strategy change trajectory and business domain tail event data in the user's historical risk assessment records, quantify the semantic change degree of the user's risk preference based on the strategy change trajectory, and simulate the asset risk resonance effect of the user's investment portfolio based on the business domain tail event data; A risk report generation module is used to optimize the multidimensional risk assessment framework by combining the fund risk coupling index, the semantic change degree and the asset risk resonance effect to obtain a target assessment framework, generate a risk hotspot distribution map of the user based on the dynamic risk transmission system, and generate a dynamic risk topology report of the user based on the risk hotspot distribution map and the target assessment framework.

Images