Automatically updated intelligent risk control system

Abstract

The invention relates to the technical field of intelligent risk control, in particular to an automatically updated intelligent risk control system, which comprises a key factor analysis module, a dynamic probability mapping module, a distribution weight adjustment module and a risk control rule analysis module. According to the method, accurate analysis of core financial risk control factor association levels and scientific classification of secondary financial risk control factors are realized by analyzing the interaction strength, the fluctuation amplitude and the trigger threshold range of the financial risk control factors, the recognition capability of complex risk relationships and influence chains is improved, and the risk management efficiency is improved through a dynamic dependency relationship structure. The method comprises the following steps: optimizing trigger association analysis, enhancing the adaptive capacity to a dynamic risk scene, combining a weight attribution value and dependency analysis, optimizing factor distribution characteristics, enhancing the regulation and control capacity to risk factor priorities, dynamically adjusting the relationship between financial risk control factors and risk control deployment, and improving the real-time evaluation and regulation and control capacity to risk events. Compared with the prior art, the precision, the flexibility and the efficiency are remarkably improved.

Description

Technical Field

[0001] The present invention relates to the field of intelligent risk control technology, and in particular to an automatically updated intelligent risk control system. Background Art

[0002] The field of intelligent risk control technology includes technologies that use computer systems and intelligent algorithms to predict, evaluate and control risks. Its core content is based on large-scale data analysis and processing, and uses intelligent algorithms and models to dynamically evaluate potential risks and respond to and manage related risk events in real time. Intelligent risk control technology as a whole covers data collection, analysis, modeling, risk prediction and decision support, and is usually used in industries such as finance, insurance, and supply chain management. The systematic characteristics of this field lie in the ability to quickly process data, dynamically perceive risks, and make decisions through automation technology.

[0003] Among them, the automatically updated intelligent risk control system refers to a technical solution that can automatically detect and obtain external risk change data, and dynamically adjust the system risk control strategy based on a preset model or algorithm. The patent subject is aimed at the problem that the risk control system needs to be frequently updated and adapted to dynamic risk scenarios. It covers the automatic capture and classification of real-time external data by the data acquisition module, the risk identification and classification functions based on the specified rule base, the completion of risk level assessment in combination with model training, and the process of generating updated strategies through automatic decision-making algorithms. Through technical means, the system can realize the automatic adjustment and upgrade of risk control strategies without human intervention.

[0004] The shortcomings of existing technologies in the field of risk control are mainly manifested in the lack of dynamic perception and real-time update capabilities. Although they can collect data and model and analyze risks, their hierarchical processing of financial risk control factors is relatively rough, and it is difficult to accurately identify the correlation hierarchy and dependency characteristics between core and secondary financial risk control factors, resulting in insufficient control over key influencing factors in complex risk scenarios. Existing technologies lack flexibility in the dynamic adjustment of conditional trigger probabilities, and the analysis of the fluctuation range of trigger probabilities is insufficient, and it is impossible to form an effective mapping of conditional relationships between factors, so it is difficult to adapt to changes in trigger conditions in dynamic risk scenarios. In terms of weight distribution management, existing technologies use a static weight allocation method, which cannot be dynamically adjusted according to changes in the priority of risk factors, resulting in insufficient response capabilities of risk control systems to sudden or variable risk events. In terms of financial risk control factor classification and rule analysis, most existing technologies perform static risk grading based on a preset rule base, lack support for dynamic dependencies and real-time adjustment capabilities between financial risk control factors, and thus limit the accuracy and adaptability of risk control strategies, resulting in the system being prone to lags and inefficiencies in dynamic scenarios, and it is difficult to meet the requirements of modern risk control for efficient dynamic decision support. Summary of the invention

[0005] The purpose of the present invention is to solve the shortcomings of the prior art and to propose an automatically updated intelligent wind control system.

[0006] In order to achieve the above-mentioned purpose, the present invention adopts the following technical solution: an automatically updated intelligent risk control system comprises: The key factor analysis module collects risk data of market price fluctuations, extracts financial risk control factors from risk data, extracts interactive data and analyzes the correlation network of financial risk control factors, refines the data classification structure to identify core financial risk control factors, decomposes data hierarchies to determine secondary financial risk control factors, classifies and separates original data features, extracts core hierarchies, merges secondary factor information to optimize hierarchical relationships, and generates a hierarchical structure feature group of financial risk control factors; The dynamic probability mapping module analyzes the conditional triggering probability and weight fluctuation range of the primary and secondary financial risk control factors based on the hierarchical structure feature group of the financial risk control factors, maps the conditional relationship between the primary and secondary factors in combination with the interaction intensity and the trigger probability, generates a factor dynamic probability map, adjusts the trigger probability, screens and quantifies the dynamic dependency relationship between the primary and secondary financial risk control factors, and obtains dynamic conditional dependency structure data; The distribution weight adjustment module extracts the dynamic weight balance value and attribution range of the primary and secondary financial risk control factors based on the dynamic conditional dependency structure data, adjusts the weight distribution between the primary and secondary financial risk control factors, generates a dynamic factor weight data set, analyzes the factor interaction intensity and weight fluctuation trend in combination with the dynamic conditional dependency structure, reallocates the factor weights and optimizes the weight attribution value of the primary financial risk control factor, analyzes the influencing factors and the correlation dependency, and generates an optimized factor distribution feature map; Based on the optimized factor distribution characteristic diagram, the risk control rule parsing module parses the dependency weights and dynamic relationship rules of primary and secondary financial risk control factors, generates a rule set between financial risk control factors, classifies each financial risk control factor, evaluates the influence, adjusts the frequency amplitude and sequence parameters, and generates an automated risk control dynamic adjustment plan.

[0007] As a further solution of the present invention, the steps for obtaining the hierarchical structure feature group of financial risk control factors are specifically as follows: Based on the interaction strength, original fluctuation range and trigger threshold range of financial risk control factors, the interaction strength value is calculated, and the financial risk control factor pairs with interaction strength greater than the threshold are screened. The financial risk control factor pairs are assigned association weights through the strength ratio to generate a financial risk control factor interaction relationship table; Based on the financial risk control factor interaction table, the volatility data of the associated factors are extracted, the extreme value difference in the time series is calculated and compared with the volatility threshold range, and the financial risk control factors that exceed the range are counted to obtain the dependency characteristic table of the financial risk control factors; Based on the dependency characteristic table of the financial risk control factors, combined with the interaction relationship table and the fluctuation range, the hierarchy of the financial risk control factors is integrated by weighting, using the formula: ; Calculate the hierarchical weights of financial risk control factors and generate a hierarchical structure feature group of financial risk control factors; in, Represents the stratified weights of financial risk control factors, represents the interaction weight coefficient of each factor, Represents the volatility of financial risk control factors, is the mean of the factor fluctuation, is the standard deviation of the factor volatility, Represents the number of financial risk control factors, is the factor index.

[0008] As a further solution of the present invention, the step of obtaining the factor dynamic probability map is specifically as follows: Based on the hierarchical structure feature group of the financial risk control factors, the trigger probability of the primary and secondary factors is extracted, the fluctuation range of the primary factor and the dependent factor set are called, and the formula is adopted: ; Calculate the conditional trigger probability and generate the conditional trigger probability results of financial risk control factors; in, represents the probability of condition triggering, For the The fluctuation value of a factor, is the fluctuation mean, is the weight value of the dependency factor, is the standard deviation of the fluctuation value, is the number of dependent factors; Based on the trigger probability results of the financial risk control factor conditions, the weight change trend is extracted through the fluctuation range of the primary and secondary factor weights, the time series interval is divided, the fluctuation interval is extracted, and the factor weight association table is generated in combination with the trigger probability results; Based on the factor weight association table, the conditional relationship between primary and secondary factors is mapped, the dynamic probability of primary and secondary factors is analyzed, the weight fluctuation and trigger probability mapping are adjusted, the joint probability value is extracted, and the factor dynamic probability map is generated.

[0009] As a further solution of the present invention, the step of acquiring the dynamic condition-dependent structure data is specifically as follows: Based on the dynamic probability map of the factors, the trigger conditions of the financial risk control factors are analyzed, the trigger probability values ​​are associated, the probability range of the trigger conditions is adjusted, the impact of the trigger conditions is quantified, and a trigger probability adjustment data set is generated; Based on the trigger probability adjustment data set, sort the trigger probability of each financial risk control factor, screen the probability values ​​of the financial risk control factors, analyze the dependency conditions of the primary and secondary factors, analyze the overlapping range and effect strength of the dependency conditions, and generate a dependency table of the primary and secondary financial risk control factors; Based on the conditional dependency table of primary and secondary financial risk control factors, the conditional dependency paths are analyzed item by item, the dynamically changing core data points are extracted, the path information dependencies are sorted out, the data point combination method is adjusted, and the dynamic conditional dependency structure data is obtained.

[0010] As a further solution of the present invention, the steps of acquiring the dynamic factor weight data set are specifically as follows: According to the dynamic condition dependency structure data, the weight values ​​and correlation ranges of the primary and secondary financial risk control factors are extracted, the degree of factor dependence and the amplitude of change in the dynamic condition dependency structure are analyzed, and the correlation weight distribution data of the primary and secondary financial risk control factors are obtained; The data on the weight distribution of the primary and secondary financial risk control factors are normalized, and the risk contribution rate of the factors is adjusted according to the attribution scope, using the formula: ; Generate dynamic weighted balance values ​​of factors within the attribution range; in, Indicates The dynamic weight balance value of financial risk control factors, For the The degree of dependence on financial risk control factors, and is the dynamic adjustment coefficient, Represents the number of total financial risk control factors; According to the dynamic weight balance value within the attribution range and combined with the adjustment results of the primary and secondary financial risk control factors, the weight distribution of the primary and secondary financial risk control factors is optimized to determine whether the dynamic factor weight balance value meets the dynamic stability condition. If so, a dynamic factor weight data set is generated.

[0011] As a further solution of the present invention, the step of obtaining the optimized factor distribution characteristic graph is specifically as follows: Based on the dynamic factor weight data set, the original weights of the main financial risk control factors are classified by category, the weight data are extracted, and normalized, the weight proportions of the factors in each category are analyzed, the weight differences between categories are extracted, and the initial weights of the main financial risk control factors are obtained; Based on the initial weights of the main financial risk control factors, decompose the total weights of the factors in each category, assign corresponding weights to the cross-category associated factors, adjust the distribution values ​​within the categories, analyze the total category weights and global weight distribution, and obtain the adjusted factor attribution weight distribution results; Based on the adjusted factor attribution weight distribution results, the correlation data between factors are extracted, the mutual influence under the factor dependency relationship is identified, and the dependency and correlation strength are numerically analyzed to generate an optimized factor distribution characteristic diagram.

[0012] As a further solution of the present invention, the steps of obtaining the rule set between the financial risk control factors are specifically as follows: Based on the optimized factor distribution characteristic diagram, the interaction intensity, dependency and trigger threshold information between the primary and secondary financial risk control factors are extracted, the trigger frequency and fluctuation range of the primary financial risk control factors are analyzed, and the correlation of the primary financial risk control factors is evaluated in combination with the dependency of the secondary financial risk control factors in differentiated scenarios, so as to generate the interaction relationship characteristics of the primary and secondary financial risk control factors; Analyze the interactive relationship characteristics of the primary and secondary financial risk control factors, and model the dynamic relationship rules of each pair of financial risk control factors using the formula: ; Calculate the rule strength values ​​between financial risk control factors and generate a set of associated features of financial risk control factors; in, Indicates the rule strength value between financial risk control factors, represents the interaction frequency of financial risk control factors, Indicates the criticality weight of financial risk control factors, Indicates the fluctuation range of financial risk control factors, Indicates the trigger threshold, represents the correlation adjustment coefficient; Based on the financial risk control factor association feature set, the primary and secondary financial risk control factor rule strength data is called, the factor relationship characteristics in the rules are grouped and analyzed, the financial risk control factors are sorted and classified according to the rule strength, the rules in the primary and secondary factor relationships are feature-labeled, and a rule set between financial risk control factors is generated.

[0013] As a further solution of the present invention, the steps for obtaining the automatic risk control dynamic adjustment plan are specifically as follows: Based on the rule set between the financial risk control factors, analyze the category characteristics of the financial risk control factors, extract differentiated category factor data, evaluate factor weights, analyze the contribution ratio of factors within the category, and obtain the classification and weight evaluation results of financial risk control factors; Based on the classification and weight assessment results of the financial risk control factors, extract the factor distribution data within the category, adjust the cross-category factor distribution structure, redistribute the total amount of the category, identify the factor contribution value according to the weight change, improve the weight relationship between categories, and obtain the influence and distribution optimization results of the financial risk control factors; Based on the influence and distribution optimization results of the financial risk control factors, the factor sequence data and frequency amplitude are extracted, the weight distribution range of the associated factors is analyzed, the weight change value is adjusted, the global dynamic parameters are reallocated, the dynamic distribution data is integrated, and an automated risk control dynamic adjustment plan is constructed.

[0014] Compared with the prior art, the advantages and positive effects of the present invention are: In the present invention, by analyzing the interaction intensity, original fluctuation amplitude and trigger threshold range of financial risk control factors, accurate analysis of the correlation level of core financial risk control factors is achieved, and secondary financial risk control factors are scientifically classified and layered, the dependency characteristics of financial risk control factors are effectively extracted, and the ability to identify the complexity of the relationship and the impact chain of financial risk control factors is significantly improved. Combined with the conditional trigger probability and weight fluctuation range of primary and secondary financial risk control factors, a dynamic dependency relationship structure between primary and secondary factors is formed through dynamic mapping of interaction intensity and trigger probability, the trigger correlation analysis between financial risk control factors is optimized, and the adaptability to dynamic changes in risk scenarios is improved. In the weight distribution adjustment link, the weight attribution range and dynamic balance value of financial risk control factors are accurately extracted, and the primary and secondary financial risk control factors are realized. The optimization of the weight distribution among risk control factors significantly strengthens the management and regulation capabilities of risk factor priorities. Combined with the weight attribution values ​​and correlation dependency analysis of the main financial risk control factors, the distribution characteristic diagram of the factors is further optimized, effectively enhancing the ability to interpret the distribution characteristics of financial risk control factors and their interactions. In the rule parsing and dynamic adjustment links, the dynamic adjustment and optimized deployment of financial risk control factors are realized through the analysis of dependent weights, dynamic relationships and classification rules, and the rapid response capability of risk control strategies to changing risk scenarios is enhanced. By gradually optimizing the stratification, correlation, weight distribution and dynamic relationships of financial risk control factors, the real-time assessment, dynamic regulation and strategy optimization capabilities of risk events are comprehensively improved, achieving significant gains in accuracy, flexibility and efficiency compared to existing technologies. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1 is a system flow chart of the present invention; Figure 2 It is a flow chart of the hierarchical structure feature group of financial risk control factors in the present invention; Figure 3 It is a flow chart of the dynamic probability map of factors in the present invention; Figure 4 A flow chart of dynamic conditional dependent structure data in the present invention; Figure 5 A flow chart of a dynamic factor weight data set in the present invention; Figure 6 It is a flow chart of the factor distribution characteristic diagram after optimization in the present invention; Figure 7It is a flow chart of the rule set between financial risk control factors in the present invention; Figure 8 The figure is a flow chart of the automatic risk control dynamic adjustment scheme in the present invention. DETAILED DESCRIPTION

[0016] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. 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.

[0017] In the description of the present invention, it should be understood that the terms "length", "width", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside" and the like indicate positions or positional relationships based on the positions or positional relationships shown in the drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as limiting the present invention. In addition, in the description of the present invention, "multiple" means two or more, unless otherwise clearly and specifically defined.

[0018] See also Figure 1 , an automatically updated intelligent risk control system includes: The key factor analysis module collects risk data of market price fluctuations, extracts financial risk control factors from risk data, extracts interactive data and analyzes the correlation network of financial risk control factors, refines the data classification structure to identify core financial risk control factors, decomposes data hierarchies to determine secondary financial risk control factors, classifies and separates original data features, extracts core hierarchies, merges secondary factor information to optimize hierarchical relationships, and generates a hierarchical structure feature group of financial risk control factors; The dynamic probability mapping module analyzes the conditional triggering probability and weight fluctuation range of the primary and secondary financial risk control factors based on the hierarchical structure feature group of the financial risk control factors, maps the conditional relationship between the primary and secondary factors in combination with the interaction intensity and the trigger probability, generates a factor dynamic probability map, adjusts the trigger probability, screens and quantifies the dynamic dependency relationship between the primary and secondary financial risk control factors, and obtains dynamic conditional dependency structure data; The distribution weight adjustment module extracts the dynamic weight balance value and attribution range of the primary and secondary financial risk control factors based on the dynamic conditional dependency structure data, adjusts the weight distribution between the primary and secondary financial risk control factors, generates a dynamic factor weight data set, analyzes the factor interaction intensity and weight fluctuation trend in combination with the dynamic conditional dependency structure, reallocates the factor weights and optimizes the weight attribution value of the primary financial risk control factor, analyzes the influencing factors and the correlation dependency, and generates an optimized factor distribution feature map; Based on the optimized factor distribution characteristic diagram, the risk control rule parsing module parses the dependency weights and dynamic relationship rules of primary and secondary financial risk control factors, generates a rule set between financial risk control factors, classifies each financial risk control factor, evaluates the influence, adjusts the frequency amplitude and sequence parameters, and generates an automated risk control dynamic adjustment plan.

[0019] The hierarchical structure feature group of financial risk control factors includes association hierarchy, classification of secondary financial risk control factors, hierarchical information, and dependency characteristics. The dynamic probability map of factors includes conditional trigger probability, weight fluctuation range, and conditional relationship between primary and secondary factors. The dynamic conditional dependency structure data includes dynamic dependency relationship, trigger probability adjustment, and conditional dependency structure. The dynamic factor weight data set includes dynamic weight balance value, weight attribution range, and weight distribution optimization data. The optimized factor distribution feature map includes the weight attribution value of the main financial risk control factor, the influence of the financial risk control factor, and the associated dependency characteristics. The rule set between financial risk control factors includes dependency weight rules, dynamic relationship rules, and financial risk control factor classification rules. The automated risk control dynamic adjustment plan includes financial risk control factor influence assessment, frequency amplitude optimization, and adjustment sequence parameters.

[0020] See also Figure 2 ,The specific steps for obtaining the hierarchical structure feature group of financial risk control factors are as follows: Based on the interaction strength, original fluctuation range and trigger threshold range of financial risk control factors, the interaction strength value is calculated, and the financial risk control factor pairs with interaction strength greater than the threshold are screened. The financial risk control factor pairs are assigned association weights through the strength ratio to generate a financial risk control factor interaction relationship table; First, it is necessary to call the interaction data set between the core financial risk control factors, extract the interaction parameters of each core financial risk control factor and its associated factor pair, use the relative impact factor ratio calculation formula for the interaction parameters of each pair of factors, take the ratio of the maximum and minimum values ​​of the interaction value between each pair of financial risk control factors, and then normalize the comparison value. Next, call the set interaction strength threshold, and compare all interaction data with the threshold one by one. Through comparative analysis, all factor pairs with interaction strength ratios greater than the threshold are screened out. Finally, based on the screening results, each pair of strength factors is normalized and assigned their interaction weights to form a financial risk control factor interaction relationship table.

[0021] Based on the interaction table of financial risk control factors, the volatility data of related factors are extracted, the extreme value difference in the time series is calculated and compared with the volatility threshold range, and the financial risk control factors that exceed the range are counted to obtain the dependency characteristic table of financial risk control factors; The volatility parameter values ​​of each financial risk control factor are extracted according to the time series, and the actual volatility of each factor is obtained by calculating the difference between the highest and lowest points in the continuous data. The calculated results are compared with the set volatility thresholds one by one, and the upper and lower limits of the thresholds are compared to determine whether the volatility exceeds the range. For factors whose volatility exceeds the range, the time point and change trend are recorded, and the factors are classified as significant volatility factors according to the trend direction (for example, when the volatility continues to increase). Further, the dependency data related to the core factors are obtained through correlation analysis. The data includes the dependency time, volatility increase, and the intensity change value of the correlation factor, and the dependency characteristic table of the financial risk control factors is obtained.

[0022] Based on the dependency characteristic table of financial risk control factors, combined with the interaction relationship table and volatility, the hierarchy of financial risk control factors is integrated through weighting, using the formula: ; Calculate the hierarchical weights of financial risk control factors and generate a hierarchical structure feature group of financial risk control factors; in, Represents the stratified weights of financial risk control factors, represents the interaction weight coefficient of each factor, Represents the volatility of financial risk control factors, is the mean of the factor fluctuation, is the standard deviation of the factor volatility, Represents the number of financial risk control factors, is the factor index; The benefit of the formula is that, by introducing the interaction weight coefficient and the factor volatility standardization parameter, it effectively combines the interaction intensity and the comprehensive impact of the volatility of financial risk control factors, reflecting the key role of the correlation between financial risk control factors in the overall hierarchical structure; It represents the final weight of the financial risk control factor stratification, and is a comprehensive indicator calculated by combining the volatility of the financial risk control factors and the interaction weight. Representative The weight parameter of each factor in the overall interaction relationship is calculated by normalization. Representative The fluctuation range of a factor is obtained by calculating the difference between the maximum and minimum values ​​in the time series of the factor. For the The average fluctuation range of the factors, is the standard deviation of the factor’s volatility, is the total number of financial risk control factors, is the factor index, and the formula is completed by calculating and adding each factor one by one; Assume that there are three financial risk control factors, and their interaction weights are , the corresponding fluctuation range is The average fluctuation ranges are The standard deviation of the fluctuation range is ; The calculation process is as follows: For the first factor: ; For the second factor: ; For the third factor: ; Substituting the above results into the total weight formula: ; This result shows that the calculated stratification weight value , combined with the stratified weight values, can be further used to optimize and adjust the stratified structure of financial risk control factors, and serve as the key basic result for subsequent calculations and analysis.

[0023] See also Figure 3 , the specific steps for obtaining the factor dynamic probability map are: Based on the hierarchical structure feature group of financial risk control factors, the trigger probability of primary and secondary factors is extracted, the fluctuation range of the primary factor and the set of dependent factors are called, and the formula is used: ; Calculate the conditional trigger probability and generate the conditional trigger probability results of financial risk control factors; in, represents the probability of condition triggering, For the The fluctuation value of a factor, is the fluctuation mean, is the weight value of the dependency factor, is the standard deviation of the fluctuation value, is the number of dependent factors; The formula is beneficial in that it calculates the conditional trigger probability by comprehensively calculating the fluctuation standard deviation and weight of the dependent factors, and combines the fluctuation relationship and intensity influence between the primary and secondary factors to optimize the dynamic performance of the conditional trigger probability and enhance the multi-dimensional dynamic analysis ability of the primary factor on the dependent factor. In the formula, The conditional trigger probability of the primary and secondary financial risk control factors is calculated by dynamically analyzing the ratio of the fluctuation range and standard deviation between the primary and secondary factors. For the The fluctuation value of each dependent factor is obtained by monitoring the fluctuation range of the main factor under the dynamic change of its dependent factor. The monitoring process includes real-time data collection of the dependent factor and extraction of benchmark data at basic time points. is the mean of the fluctuation value of the dependency factor, which is calculated by the fluctuation values ​​at multiple sampling time points. is the weight value of the dependent factor. The weight value is normalized according to the correlation strength of the dependent factor to the main factor. is the standard deviation of the volatility of the dependent factor, which is calculated by the degree of dispersion of the volatility of the dependent factor time series. is the total number of all dependent factors. Through the interactive analysis of dependent factors, the summation symbol represents the accumulation of the calculation results of each factor. The overall formula combines the fluctuation range of dependent factors with the fluctuation mean of the main factor, providing a numerical basis for the dynamic correlation of the trigger probability. Assume that there are three dependent factors, and their fluctuation values ​​are , the corresponding mean is The weight values ​​are , the standard deviation is ; Molecular calculation: ; Denominator calculation: ; Conditional trigger probability calculation: ; The result shows that the conditional trigger probability is 0.464, indicating that the triggering intensity of the main factor on the dependent factor under dynamic conditions is at a medium to low level, which can be further used as the basic quantitative basis for the analysis of the conditional relationship between the main and secondary factors.

[0024] Based on the trigger probability results of financial risk control factors, the weight change trend is extracted through the fluctuation range of primary and secondary factor weights, the time series interval is divided, the fluctuation interval is extracted, and the factor weight association table is generated in combination with the trigger probability results; First, it is necessary to divide the weight data of each primary and secondary factor into time periods. By calculating the mean of the weight data according to a fixed time window, the change trend of the weight data is extracted and the fluctuation range of each time period is calculated. The average weight of each weight fluctuation value in the time series is subtracted from the average weight of the time period to obtain the weight change rate of the time period. The weight change rate of each time period is further classified to determine whether the change rate exceeds the set threshold range. For example, the weight change rate that exceeds the range can be defined as a significant fluctuation point, while the weight change rate that is below the range can be regarded as a stable fluctuation point. Through the classification and statistics of the weight change rate, the fluctuation data feature table of the primary and secondary factors is further extracted, and the feature table is matched with the conditional trigger probability data one by one to generate the correlation relationship between the fluctuation of the primary and secondary factors and the trigger probability. The factor weight association table is generated through the matched correlation relationship. This table combines the fluctuation characteristics of the primary and secondary factors with the conditional trigger strength, providing basic support for the subsequent mapping of dynamic relationships.

[0025] Based on the factor weight association table, map the conditional relationship between primary and secondary factors, analyze the dynamic probability of primary and secondary factors, adjust the weight fluctuation and trigger probability mapping, extract the joint probability value, and generate the factor dynamic probability map; First, it is necessary to sample the dynamic probabilities of primary and secondary factors, perform quadratic fitting on the sampled data according to the time window, calculate the changing trend of the dynamic probability value through the fitting formula, reset the time window interval for the fitted dynamic probability value, correct the dynamic probability value of each time window one by one, and perform joint probability calculation on the dynamic probability value of each time window with the data of trigger probability and weight fluctuation range to obtain the joint probability distribution of primary and secondary factors. Through the calculated joint probability distribution, the dynamic relationship between primary and secondary factors is visualized, and a probability change trend graph is drawn to map the dynamic conditional relationship between primary and secondary factors. All the drawn dynamic relationship trend graphs are integrated to finally form a factor dynamic probability map, which combines the conditional trigger probability and weight fluctuation range data of primary and secondary factors to provide intuitive expression and quantitative support for the analysis of dynamic financial risk control factors.

[0026] See also Figure 4 ,The specific steps for obtaining dynamic conditional dependency structure data are: Based on the dynamic probability map of factors, analyze the trigger conditions of financial risk control factors, associate the trigger probability values, adjust the probability range of trigger conditions, quantify the impact of trigger conditions, and generate a trigger probability adjustment data set; First, each financial risk control factor in the factor dynamic probability map needs to be defined as an independent node, and the trigger condition relationship between it and other factors needs to be gradually established. By analyzing the probability distribution characteristics of each node one by one, the key parameters required for the trigger condition are clarified. For example, when the probability of a node exceeds a certain threshold, the condition changes of the triggering associated factor. The trigger condition needs to be refined and decomposed into smaller probability range intervals. By calculating the actual trigger probability value of the associated condition, the specific impact of the trigger condition on each interval is quantified. According to the quantified result, the probability interval of the trigger condition is adjusted to make the adjusted conditional probability closer to the actual needs of risk management. For example, by redistributing weights, the trigger probability of certain conditions is limited to between 0.2 and 0.4, so as to avoid low probability conditions from causing too much interference to the results in risk analysis, and generate a trigger probability adjustment data set, which contains the trigger probability value of each financial risk control factor under different condition ranges for subsequent dynamic analysis and screening.

[0027] Adjust the data set based on the trigger probability, sort the trigger probability of each financial risk control factor, screen the probability value of the financial risk control factor, analyze the dependency conditions of the primary and secondary factors, analyze the overlapping range and effect strength of the dependency conditions, and generate the dependency table of the primary and secondary financial risk control factors; Analyze and screen the trigger probability values ​​in turn, classify the probability values ​​in order of size and mark the characteristics of primary and secondary factors. In the sorting process, it is necessary to group each factor based on the actual impact of the trigger conditions, extract the primary and secondary factors with greater impact, analyze the dependency conditions between each financial risk control factor, calculate the dependency condition range of each financial risk control factor item by item, and analyze the overlapping parts within the range to determine the intensity of the financial risk control factor under different dependency conditions. For example, by calculating the overlapping range of multiple factors for the same trigger condition, the proportion of a factor in the dependency condition is clarified, so as to accurately measure its impact on the overall risk and generate a primary and secondary financial risk control factor condition dependency table. The table records the degree of dependence of each factor on different conditions and its interaction with the factor, providing a basis for subsequent path analysis.

[0028] Based on the condition dependency table of primary and secondary financial risk control factors, analyze the condition dependency path item by item, extract the core data points of dynamic changes, sort out the path information dependency relationship, adjust the data point combination method, and obtain dynamic condition dependency structure data; The dependency paths of each primary and secondary factor are gradually extracted to form a path chain with time or state changes as the core, and the core data points with dynamic changes in each dependency path are extracted. By screening the nodes with larger changes, the dynamic adjustment rules of the core data points are locked, and the main nodes in the dependency path and their associations are sorted out. When sorting out the path information, the dependency strength of each path is cross-compared, and the combination method of each core data point is adjusted to ensure that the dynamic dependency path can be efficiently updated in the risk control system. In the process of adjusting the data point combination method, the rule range is set for the nodes that frequently fluctuate in dynamic changes, and the combination method is gradually optimized through multiple verifications to obtain dynamic conditional dependency structure data. This data can automatically adapt to the real-time update requirements in the risk control system, thereby further improving the system's intelligent processing capabilities.

[0029] See also Figure 5 , the steps to obtain the dynamic factor weight dataset are as follows: According to the dynamic conditional dependency structure data, the weight values ​​and correlation ranges of the primary and secondary financial risk control factors are extracted, the factor dependence degree and change range in the dynamic conditional dependency structure are analyzed, and the correlation weight distribution data of the primary and secondary financial risk control factors are obtained; The dynamic conditional dependency structure data is derived from the observed results of multi-variable interdependence in the actual environment. It is necessary to calculate the dependency and mutual influence of each variable through time series analysis or conditional probability distribution. The process of extracting weight values ​​can be achieved through multiple conditional judgments and correlation analysis. The dependency matrix between variables is constructed based on the dynamic conditional dependency structure data. The construction of this matrix is ​​based on the conditional probability distribution between variables. The quantitative value of the dependency is obtained through normalized calculation, and the correlation range of primary and secondary financial risk control factors is analyzed. The key factors and secondary factors are screened by setting the dependency threshold. Factors above the threshold are determined as primary financial risk control factors, and factors below the threshold are classified as secondary financial risk control factors. At the same time, the historical data of financial risk control factors are combined for distribution fitting, and the attribution range is further corrected. The dependency matrix and the attribution range correction results are combined to finally obtain the correlation weight distribution data of primary and secondary financial risk control factors.

[0030] Normalize the distribution data of the associated weights of the primary and secondary financial risk control factors, and adjust the risk contribution rate of the factors according to the attribution scope using the formula: ; Generate dynamic weighted balance values ​​of factors within the attribution range; in, Indicates The dynamic weight balance value of financial risk control factors, For the The degree of dependence on financial risk control factors, and is the dynamic adjustment coefficient, Represents the number of total financial risk control factors; By introducing a dynamic adjustment factor and , can dynamically adjust the weight distribution according to the contribution of individual financial risk control factors and the overall balance, and optimize the influence distribution of financial risk control factors on a global scale; : No. The dynamic weight balance value of financial risk control factors; : No. The degree of dependence of each financial risk control factor is specifically extracted from the numerical value of the dependence matrix in the dynamic conditional dependence structure; : Individual balance weight adjustment coefficient, which is set according to the degree of deviation of the trend of primary and secondary factors in historical data; : The global weight balance influence coefficient is set according to the ratio of the number of primary and secondary factors and the stability of their weight distribution; Through the conditional probability distribution extraction of the dependency matrix, each value in the dependency matrix represents the degree of dependence between financial risk control factors, which can be calculated by the probability of actual observation data. The specific method is as follows: ; in express and The joint probability distribution of for The marginal probability of, the probability data comes from long-term observation records and statistical analysis; and It is determined by calculating the ratio of the number of primary and secondary factors (for example, the ratio of the number of primary factors to secondary factors is 3:7) and the mean and variance of historical deviation values, for example: , ; in and are the mean and standard deviation of the historical deviation values ​​of the main factors, and is the mean and standard deviation of the secondary factor deviations; Let dependency matrix extract value , , , calculate in sequence: ; ; ; Calculated dynamic weight balance value , , It represents the dynamic impact weight distribution of primary and secondary financial risk control factors in the overall structure. Factors corresponding to higher weight values ​​have higher weights in systemic risk. This distribution can provide a basis for the dynamic balance adjustment of factors within the subsequent attribution range.

[0031] According to the dynamic weight balance value within the attribution range, combined with the adjustment results of the primary and secondary financial risk control factors, the weight distribution of the primary and secondary financial risk control factors is optimized to determine whether the dynamic factor weight balance value meets the dynamic stability condition. If so, a dynamic factor weight data set is generated; By calculating the volatility index of the financial risk control factor within the attribution range, the difference between the volatility value and the dynamic weight is used as the standard for judging stability. The volatility index can be calculated by the standard deviation or coefficient of variation of historical data. For example, take the dynamic weight value sequence in the first seven periods, calculate its standard deviation as the stability criterion, screen the weight distribution of factors with lower volatility, and use this as a reference to adjust the dynamic weight of high volatility factors. The weight distribution is gradually optimized by adjusting the upper and lower limits of the risk weight within its attribution range. The specific optimization method can be achieved through linear interpolation or weight normalization. The overall risk contribution within the attribution range is judged according to the optimized dynamic weight result. If it meets the stability criterion, the optimization is terminated. Otherwise, the high volatility factor is further adjusted to finally generate a dynamic factor weight data set.

[0032] See also Figure 6 , the steps for obtaining the optimized factor distribution characteristic graph are as follows: Based on the dynamic factor weight data set, the original weights of the main financial risk control factors are grouped by category, the weight data are extracted, and normalized. The weight ratio of each category is analyzed, the weight differences between categories are extracted, and the initial weights of the main financial risk control factors are obtained. The weight of each factor is sorted according to its classification. By screening out financial risk control factors with higher weights, the distribution characteristics of primary and secondary factors in each category are clarified, the factor weights under each category are accumulated to obtain the total weight within the category, and the proportion of each factor in the category is calculated. At the same time, the proportion values ​​are normalized to unify the weight quantification standard. During the processing, the normalized data is analyzed, the internal structure characteristics of each category factor are clarified, and the degree of difference in weight data is measured based on the aggregation results between different categories. For example, the weight difference between the maximum and minimum proportion factors in each category is calculated, the weight fluctuation range across categories is extracted, and the initial weight of the main financial risk control factor is obtained. This result provides basic data support for subsequent weight decomposition and adjustment, and can significantly improve the adaptation efficiency of the dynamic factor weight data set in the risk control system.

[0033] Based on the initial weights of the main financial risk control factors, decompose the total weight of factors in each category, assign corresponding weights to cross-category related factors, adjust the distribution value within the category, analyze the total category weight and global weight distribution, and obtain the adjusted factor attribution weight distribution results; First, the total weight in each category is divided into multiple detailed intervals. According to the normalized weight ratio data, each interval is allocated to different factors in the category. At the same time, the weight requirements of cross-category related factors are given priority in the allocation process. By reallocating some weight resources within the category, relatively independent weight components are established for cross-category factors. When adjusting the distribution value within the category, the adjusted distribution of the weight of each category factor is analyzed one by one to ensure that the total weight of the category remains unchanged. At the same time, the global weight distribution is analyzed, and the distribution structure of cross-category factors is further optimized based on the proportion of each category factor weight in the global weight. For example, by calculating the degree of adaptation of the highest weight factor in each category to the global weight distribution, its weight ratio is adjusted to improve global consistency. Finally, the adjusted factor attribution weight distribution result is obtained. This distribution result can dynamically adapt to the actual influence range of different financial risk control factors in the risk control system to ensure the rationality and scientificity of the factor weight distribution.

[0034] Based on the adjusted factor attribution weight distribution results, extract the correlation data between factors, identify the mutual influence under the factor dependency relationship, and perform numerical analysis on the dependency and correlation strength to generate an optimized factor distribution characteristic diagram; By analyzing the associated data, we can gradually identify the degree of mutual influence between different factors under the dependency relationship. By establishing the correlation mapping before and after the factor weight adjustment, we can measure the dynamic changes of the weight adjustment on the factor dependency relationship. For example, we can calculate the impact of a main factor on its own weight fluctuation after the associated factor weight is adjusted. After the dependency relationship is identified, we can further conduct numerical analysis on the correlation strength between the factors. Combined with the calculation results of the dependency, we can clarify the key parameter range of the factor interaction. For example, based on the weight distribution ratio of multiple factors to a certain dependency condition, we can analyze its correlation strength and optimize the adjustment rules. According to the analysis results of the dependency and correlation strength, we can generate an optimized factor distribution feature map. This feature map can clearly show the dynamic distribution status of each factor under different correlation conditions, provide support for the real-time update of the risk control system, and effectively improve the system's ability to handle complex risk relationships.

[0035] See also Figure 7 , the specific steps for obtaining the rule set between financial risk control factors are: Based on the optimized factor distribution characteristic diagram, the interaction intensity, dependency and trigger threshold information between the primary and secondary financial risk control factors are extracted, the trigger frequency and fluctuation range of the primary financial risk control factors are analyzed, and the correlation of the primary financial risk control factors is evaluated in combination with the dependency of the secondary financial risk control factors in differentiated scenarios, and the interaction relationship characteristics of the primary and secondary financial risk control factors are generated; The triggering frequency of the main financial risk control factors is statistically calculated by extracting the number of triggering records of the main financial risk control factors in different time periods. The triggering time of each triggering event is associated with the actual fluctuation time point. The fluctuation range is calculated by the extreme value difference of the financial risk control factor in each triggering cycle. The triggering threshold is evaluated based on the statistical results of the triggering frequency of the main financial risk control factors and the changing trend of the fluctuation range. The primary and secondary factor combinations with significant interactive characteristics are screened, and the interactive relationship characteristics of the primary and secondary financial risk control factors are obtained through data characteristics.

[0036] Analyze the interactive relationship characteristics of primary and secondary financial risk control factors, and model the dynamic relationship rules of each pair of financial risk control factors using the formula: ; Calculate the rule strength values ​​between financial risk control factors and generate a set of associated features of financial risk control factors; in, Indicates the rule strength value between financial risk control factors, represents the interaction frequency of financial risk control factors, Indicates the criticality weight of financial risk control factors, Indicates the fluctuation range of financial risk control factors, Indicates the trigger threshold, represents the correlation adjustment coefficient; The benefit of the formula is that by combining the interaction frequency and fluctuation amplitude, balancing the comprehensive effect of the trigger threshold deviation and the dynamic adjustment coefficient, it can accurately describe the dynamic association rules between financial risk control factors and adapt to the changes in factor distribution characteristics in multiple scenarios; Interaction frequency The initial value of is extracted from the interaction strength of the primary and secondary factors and recorded as follows: the triggering frequencies of the primary factors are 10, 15, 20, 12, and 18, respectively, and the mean is calculated: ; Criticality Weight Adjust according to the degree of dependence of primary and secondary factors and set the weight value ; Fluctuation Range and trigger threshold Calculated through the fluctuation data of the main financial risk control factors, they are: , , adjustment coefficient The value of is adjusted by the context scenario and is set to ; Substitute all parameters into the formula for calculation: ; The result shows that the rule strength value between financial risk control factors is 28, indicating that the dynamic interaction correlation between primary and secondary factors has strong regularity and can be used for further classification and modeling.

[0037] Based on the associated feature set of financial risk control factors, call the strength data of the rules of primary and secondary financial risk control factors, group and analyze the factor relationship characteristics in the rules, sort and classify the financial risk control factors according to the rule strength, and mark the features of the rules in the primary and secondary factor relationships to generate a rule set between financial risk control factors; The rule strength of financial risk control factors is divided into three categories: high, medium and low according to the primary and secondary relationship. The rule strength classification is completed by statistically analyzing the strength value range and dividing the level boundaries. The rule strength of each level is grouped and labeled according to its dependency characteristics in the primary and secondary factor combination. The grouping process counts the scenario-related parameters of the primary and secondary factors, and combines the changing trend and distribution ratio of the rule strength in each scenario. The interdependence characteristics of the primary and secondary factors in the rule characteristics are integrated into the rule set, and finally a rule set between financial risk control factors is generated.

[0038] See also Figure 8 , the specific steps for obtaining the automated risk control dynamic adjustment plan are: Based on the rule set between financial risk control factors, analyze the characteristics of financial risk control factor categories, extract differentiated category factor data, evaluate factor weights, analyze the contribution ratio of factors within categories, and obtain the classification and weight evaluation results of financial risk control factors; Assign each factor to the corresponding category according to the definition of the rule set, and extract differentiated factor data between different categories by analyzing the characteristics within the category. By comparing the specific characteristics of the factors in each category one by one, such as comparing the factor trigger probability range, the complexity of the dependent conditions, and the degree of influence, the core scope of the differentiated factors is further clarified. The extracted differentiated factor data is classified and summarized, and evaluated item by item according to the internal weight distribution of the category. The factor contribution ratio in each category is compared with other factors to quantify its specific proportion in the total weight of the category. Through statistical analysis of the proportion data, the proportional relationship between high-weight factors and low-weight factors is identified. Combined with the characteristics of the rule set, the overall weight balance of the category is calculated, and finally the classification and weight evaluation results of financial risk control factors are obtained, providing basic support for subsequent cross-category distribution adjustments.

[0039] Based on the classification and weight assessment results of financial risk control factors, extract the factor distribution data within the category, adjust the cross-category factor distribution structure, redistribute the total amount of categories, identify the factor contribution value according to the weight change, improve the weight relationship between categories, and obtain the influence and distribution optimization results of financial risk control factors; By analyzing the distribution data within categories, the concentration and weight differences of factors within categories are analyzed. According to the actual influence range of cross-category factors, the distribution data is adjusted and the weights of cross-category factors are reallocated. For example, in high-weight categories, the total weights of some factors are allocated to low-weight categories with higher correlations to ensure that the balance between categories can be maintained after the total category amount is redistributed. During the adjustment process, the specific contribution value of each factor is identified according to the weight change, and the influence range of the factor on the category weight is quantified. By comparing the fluctuations before and after the change in the factor contribution value, the weight relationship between categories is improved, and the factor distribution state within the category is gradually optimized, so that the adjusted category structure is more coordinated, and the influence and distribution optimization results of financial risk control factors are obtained. This data can be used as the core basis of the dynamic risk control model to achieve accurate distribution of factors among different categories.

[0040] Based on the influence and distribution optimization results of financial risk control factors, extract the factor sequence data and frequency amplitude, analyze the weight distribution range of related factors, adjust the weight change value, reallocate the global dynamic parameters, integrate the dynamic distribution data, and build an automated risk control dynamic adjustment plan; By analyzing the arrangement of each weight data in the factor sequence, the distribution pattern of the factors in the global scope is clarified, and the change value of the cross-category factor weight is adjusted accordingly. According to the increase or decrease in the weights of different factors, the global dynamic parameters are gradually reallocated, so that the parameters after the weight change can better adapt to the risk control needs. When reallocating dynamic parameters, the factor frequency and correlation amplitude are used as key indicators, and the parameters with large dynamic changes are screened for key adjustments to ensure that the parameters remain stable after dynamic updates. By verifying the integrated dynamic distribution data and confirming the consistency of the correlation between the data, an automated risk control dynamic adjustment plan is finally constructed. This plan can optimize the system structure in real time according to the dynamic changes of financial risk control factors, achieve rapid response to complex risk environments, and enhance the intelligent processing capabilities of the risk control system in dynamic scenarios.

[0041] The above are only preferred embodiments of the present invention and are not intended to limit the present invention in other forms. Any technician familiar with the profession may use the technical contents disclosed above to change or modify them into equivalent embodiments with equivalent changes and apply them to other fields. However, any simple modification, equivalent change and modification made to the above embodiments based on the technical essence of the present invention without departing from the technical solution of the present invention still falls within the protection scope of the technical solution of the present invention.

Claims

1. An automatically updated intelligent risk control system, characterized in that: The system comprises: The key factor analysis module collects risk data of market price fluctuations, extracts financial risk control factors from risk data, extracts interactive data and analyzes the correlation network of financial risk control factors, refines the data classification structure to identify core financial risk control factors, decomposes data hierarchies to determine secondary financial risk control factors, classifies and separates original data features, extracts core hierarchies, merges secondary factor information to optimize hierarchical relationships, and generates a hierarchical structure feature group of financial risk control factors; The dynamic probability mapping module analyzes the conditional triggering probability and weight fluctuation range of the primary and secondary financial risk control factors based on the hierarchical structure feature group of the financial risk control factors, maps the conditional relationship between the primary and secondary factors in combination with the interaction intensity and the trigger probability, generates a factor dynamic probability map, adjusts the trigger probability, screens and quantifies the dynamic dependency relationship between the primary and secondary financial risk control factors, and obtains dynamic conditional dependency structure data; The distribution weight adjustment module extracts the dynamic weight balance value and attribution range of the primary and secondary financial risk control factors based on the dynamic conditional dependency structure data, adjusts the weight distribution between the primary and secondary financial risk control factors, generates a dynamic factor weight data set, analyzes the factor interaction intensity and weight fluctuation trend in combination with the dynamic conditional dependency structure, reallocates the factor weights and optimizes the weight attribution value of the primary financial risk control factor, analyzes the influencing factors and the correlation dependency, and generates an optimized factor distribution feature map; Based on the optimized factor distribution characteristic diagram, the risk control rule parsing module parses the dependency weights and dynamic relationship rules of primary and secondary financial risk control factors, generates a rule set between financial risk control factors, classifies each financial risk control factor, evaluates the influence, adjusts the frequency amplitude and sequence parameters, and generates an automated risk control dynamic adjustment plan.

2. The intelligent risk control system with automatic update according to claim 1, characterized in that: The steps for obtaining the hierarchical structure feature group of financial risk control factors are specifically as follows: Based on the interaction strength, original fluctuation range and trigger threshold range of financial risk control factors, the interaction strength value is calculated, and the financial risk control factor pairs with interaction strength greater than the threshold are screened. The financial risk control factor pairs are assigned association weights through the strength ratio to generate a financial risk control factor interaction relationship table; Based on the financial risk control factor interaction table, the volatility data of the associated factors are extracted, the extreme value difference in the time series is calculated and compared with the volatility threshold range, and the financial risk control factors that exceed the range are counted to obtain the dependency characteristic table of the financial risk control factors; Based on the dependency characteristic table of the financial risk control factors, combined with the interaction relationship table and the fluctuation range, the hierarchy of the financial risk control factors is integrated by weighting, using the formula: ； Calculate the hierarchical weights of financial risk control factors and generate a hierarchical structure feature group of financial risk control factors; in, Represents the stratified weights of financial risk control factors, represents the interaction weight coefficient of each factor, Represents the volatility of financial risk control factors, is the mean of the factor fluctuations, is the standard deviation of the factor volatility, Represents the number of financial risk control factors, is the factor index.

3. The intelligent risk control system with automatic update according to claim 2, characterized in that: The steps for obtaining the factor dynamic probability map are specifically as follows: Based on the hierarchical structure feature group of the financial risk control factors, the trigger probability of the primary and secondary factors is extracted, the fluctuation range of the primary factor and the dependent factor set are called, and the formula is adopted: ； Calculate the conditional trigger probability and generate the conditional trigger probability results of financial risk control factors; in, represents the probability of condition triggering, For the The fluctuation value of a factor, is the fluctuation mean, is the weight value of the dependency factor, is the standard deviation of the fluctuation value, is the number of dependent factors; Based on the trigger probability results of the financial risk control factor conditions, the weight change trend is extracted through the fluctuation range of the primary and secondary factor weights, the time series interval is divided, the fluctuation interval is extracted, and the factor weight association table is generated in combination with the trigger probability results; Based on the factor weight association table, the conditional relationship between primary and secondary factors is mapped, the dynamic probability of primary and secondary factors is analyzed, the weight fluctuation and trigger probability mapping are adjusted, the joint probability value is extracted, and the factor dynamic probability map is generated.

4. The intelligent risk control system with automatic update according to claim 3 is characterized in that: The steps for obtaining the dynamic condition-dependent structure data are specifically as follows: Based on the dynamic probability map of the factors, the trigger conditions of the financial risk control factors are analyzed, the trigger probability values ​​are associated, the probability range of the trigger conditions is adjusted, the impact of the trigger conditions is quantified, and a trigger probability adjustment data set is generated; Based on the trigger probability adjustment data set, sort the trigger probability of each financial risk control factor, screen the probability values ​​of the financial risk control factors, analyze the dependency conditions of the primary and secondary factors, analyze the overlapping range and effect strength of the dependency conditions, and generate a dependency table of the primary and secondary financial risk control factors; Based on the conditional dependency table of primary and secondary financial risk control factors, the conditional dependency paths are analyzed item by item, the dynamically changing core data points are extracted, the path information dependencies are sorted out, the data point combination method is adjusted, and the dynamic conditional dependency structure data is obtained.

5. The intelligent risk control system with automatic update according to claim 4 is characterized in that: The steps for obtaining the dynamic factor weight data set are specifically as follows: According to the dynamic condition dependency structure data, the weight values ​​and correlation ranges of the primary and secondary financial risk control factors are extracted, the degree of factor dependence and the amplitude of change in the dynamic condition dependency structure are analyzed, and the correlation weight distribution data of the primary and secondary financial risk control factors are obtained; The data on the weight distribution of the primary and secondary financial risk control factors are normalized, and the risk contribution rate of the factors is adjusted according to the attribution scope, using the formula: ； Generate dynamic weighted balance values ​​of factors within the attribution range; in, Indicates The dynamic weight balance value of financial risk control factors, For the The degree of dependence on financial risk control factors, and is the dynamic adjustment coefficient, Represents the number of total financial risk control factors; According to the dynamic weight balance value within the attribution range, combined with the adjustment results of the primary and secondary financial risk control factors, the weight distribution of the primary and secondary financial risk control factors is optimized to determine whether the dynamic factor weight balance value meets the dynamic stability condition. If so, a dynamic factor weight data set is generated.

6. The intelligent risk control system with automatic update according to claim 5, characterized in that: The steps for obtaining the optimized factor distribution characteristic graph are specifically as follows: Based on the dynamic factor weight data set, the original weights of the main financial risk control factors are classified by category, the weight data are extracted, and normalized, the weight proportions of the factors in each category are analyzed, the weight differences between categories are extracted, and the initial weights of the main financial risk control factors are obtained; Based on the initial weights of the main financial risk control factors, decompose the total weights of the factors in each category, assign corresponding weights to the cross-category associated factors, adjust the distribution values ​​within the categories, analyze the total category weights and the global weight distribution, and obtain the adjusted factor attribution weight distribution results; Based on the adjusted factor attribution weight distribution results, the correlation data between factors are extracted, the mutual influence under the factor dependency relationship is identified, and the dependency and correlation strength are numerically analyzed to generate an optimized factor distribution characteristic diagram.

7. The intelligent risk control system with automatic update according to claim 6, characterized in that: The steps for obtaining the rule set between the financial risk control factors are specifically as follows: Based on the optimized factor distribution characteristic diagram, the interaction intensity, dependency and trigger threshold information between the primary and secondary financial risk control factors are extracted, the trigger frequency and fluctuation range of the primary financial risk control factors are analyzed, and the correlation of the primary financial risk control factors is evaluated in combination with the dependency of the secondary financial risk control factors in differentiated scenarios, so as to generate the interaction relationship characteristics of the primary and secondary financial risk control factors; Analyze the interactive relationship characteristics of the primary and secondary financial risk control factors, and model the dynamic relationship rules of each pair of financial risk control factors using the formula: ； Calculate the rule strength values ​​between financial risk control factors and generate a set of associated features of financial risk control factors; in, Indicates the rule strength value between financial risk control factors, represents the interaction frequency of financial risk control factors, Indicates the criticality weight of financial risk control factors, Indicates the fluctuation range of financial risk control factors, Indicates the trigger threshold, represents the correlation adjustment coefficient; Based on the financial risk control factor association feature set, the primary and secondary financial risk control factor rule strength data is called, the factor relationship characteristics in the rules are grouped and analyzed, the financial risk control factors are sorted and classified according to the rule strength, the rules in the primary and secondary factor relationships are feature-labeled, and a rule set between financial risk control factors is generated.

8. The intelligent risk control system with automatic update according to claim 7, characterized in that: The steps for obtaining the automatic risk control dynamic adjustment plan are specifically as follows: Based on the rule set between the financial risk control factors, analyze the category characteristics of the financial risk control factors, extract differentiated category factor data, evaluate factor weights, analyze the contribution ratio of factors within the category, and obtain the classification and weight evaluation results of financial risk control factors; Based on the classification and weight assessment results of the financial risk control factors, extract the factor distribution data within the category, adjust the cross-category factor distribution structure, redistribute the total amount of the category, identify the factor contribution value according to the weight change, improve the weight relationship between categories, and obtain the influence and distribution optimization results of the financial risk control factors; Based on the influence and distribution optimization results of the financial risk control factors, the factor sequence data and frequency amplitude are extracted, the weight distribution range of the associated factors is analyzed, the weight change value is adjusted, the global dynamic parameters are reallocated, the dynamic distribution data is integrated, and an automated risk control dynamic adjustment plan is constructed.

Images