DEMATEL-ISM-based science and technology enterprise financing guarantee risk assessment method

Abstract

The invention discloses a science and technology enterprise financing guarantee risk assessment method based on DEMATEL-ISM, and relates to the technical field of risk assessment, and the method comprises the steps: S1, obtaining a first-level risk index and each second-level risk index based on the characteristics of a corresponding enterprise and the influence factors of financing guarantee; s2, utilizing DEMATEL to convert the interdependency relationship of each influence factor in the first-level risk index and the second-level risk index into two causal groups; meanwhile, performing hierarchical division on influence factors in the system by means of ISM; and S3, analyzing and visualizing the multi-level hierarchical structure model by using the directed graph to obtain a multi-layer hierarchical directed graph. The science and technology enterprise financing guarantee risk factor multi-layer recursive directed graph is constructed, the mutual influence relation of the science and technology enterprise financing guarantee risk factors can be reflected, the structure and the process reflecting conduction among the risk factors can be obtained, and therefore the risk factors and the incidence relation can be rapidly analyzed.

Description

Technical Field

[0001] The present invention relates to the technical field of risk assessment, and in particular to a DEMATEL-ISM-based risk assessment method for financing guarantees of technology-based enterprises. Background Art

[0002] The development and upgrading of science and technology have been advancing by leaps and bounds, and technology-based enterprises have become one of the important driving forces for economic development. Data show that about 98% of technology-based enterprises are small and medium-sized enterprises. However, the characteristics of large R&D investment and long payback period of technology-based enterprises make them have "short, frequent, scattered and small" financing needs. In recent years, the innovation fields of small and medium-sized enterprises have become more extensive, and they have become an emerging force in high-tech industries such as information technology, energy conservation and environmental protection, and high-end manufacturing. If small and medium-sized technology companies do not have a good sense of risk management and lack effective financing planning, they will easily encounter difficulties in financing. For this reason, how to effectively reduce or avoid risks has become a crucial issue that affects the healthy development of the financing guarantee industry for technology-based enterprises, and has therefore become the focus of research in the industry.

[0003] After searching, a Chinese patent (publication number: CN109934701B) discloses a risk assessment method and device for equity pledge, which includes: synchronizing the underlying original data to the kudu database through the Binlog process; performing multi-table association calculations on the original data in the kudu database through ImpalaSQL to obtain an intermediate result table; reading the intermediate result table in the kudu database through the TAF process, performing risk assessment operations based on the equity pledge data in the table, and storing the risk assessment results in Redis; providing equity pledge service content through the external interface of the TAF framework so that users can obtain risk assessment results.

[0004] In the complex field of risk assessment of financing guarantee for technology-based enterprises, many influencing factors do not exist in isolation, but are interrelated and influence each other. However, existing research rarely pays attention to the interaction between the influencing factors of financing guarantee, and lacks in-depth research on the relationship between the influencing factors. Proving the correlation between the influencing factors of financing guarantee for technology-based enterprises and clarifying the interaction between the influencing factors are of great significance for improving the financing risk management system of technology-based enterprises and enhancing risk management capabilities. Therefore, this paper proposes a risk assessment method for financing guarantee for technology-based enterprises based on DEMATEL-ISM. Summary of the invention

[0005] The purpose of the present invention is to provide a risk assessment method for financing guarantee of technology-based enterprises based on DEMATEL-ISM to solve the problems mentioned in the above background technology.

[0006] The present invention can be implemented by the following technical solution: a risk assessment method for financing guarantee of technology-based enterprises based on DEMATEL-ISM, the method comprising the following steps:

[0007] S1. Based on the characteristics of the corresponding enterprise and the factors affecting its financing guarantee, obtain the influencing indicators of the security risk of the financing guarantee, establish a financing guarantee risk evaluation indicator system from the four dimensions of the corresponding enterprise's internal risk, guarantee business risk, macroeconomic risk, and industry market risk, and take the four types of risks as the first-level risk indicators, and the remaining influencing factors under the four types of risks as the second-level risk indicators, and label each first-level risk indicator and each second-level risk indicator;

[0008] S2. Use DEMATEL to transform the interdependence of each influencing factor in the primary risk index and the secondary risk index into two causal groups, and quantitatively analyze the influence degree between the system elements. By calculating the influence degree matrix between each influencing factor, an influence relationship diagram is obtained, that is, the mutual influence between each wind direction influencing factor is transformed into nodes and edges in the influence relationship diagram, and the influence relationship diagram is used to find the key influencing factors;

[0009] At the same time, ISM is used to classify the influencing factors in the system and build a bottom-up multi-level hierarchical model, so as to more clearly present the hierarchical relationship between the influencing factors;

[0010] S3. Utilize directed graph analysis and visualization of multi-level hierarchical structure models to obtain multi-layer hierarchical directed graphs, and based on the multi-layer hierarchical directed graphs, obtain a multi-layer hierarchical explanatory structure model of the influencing factors of corporate financing guarantees, and obtain the influencing mechanism of each influencing factor on financing guarantees and the transmission relationship of each influencing factor from the model;

[0011] The financing guarantee safety network is constructed according to the multi-layer hierarchical explanatory structure model. When consensus is reached at each influencing factor node, consensus is reached only between nodes that are interconnected and have a transmission relationship with the upper and lower layers.

[0012] A further technical improvement of the present invention is that the step S2 includes the following implementation steps:

[0013] Z1. Directly affects the establishment of the relationship matrix;

[0014] Determine the set of influencing factors and assign scores to the degree of influence of each influencing factor on each other. The assigned scores represent the degree of influence of influencing factor Si on influencing factor Sj.

[0015] Z2. Establishment of comprehensive impact matrix;

[0016] In order to eliminate the influence of different dimensions, according to the formula Normalize the direct influence matrix to obtain a normalized direct influence matrix B;

[0017] Among them, b ij represents the normalized direct influence, that is, the influence of influencing factor i on influencing factor j after normalization, b ij Used in direct impact matrix B;

[0018] a ij represents the influencing factors in the initial direct influence matrix, specifically the degree of influence of influencing factor i on influencing factor j, which is scored by experts and scholars in this embodiment;

[0019] Indicates the total influence of influencing factor i on other influencing factors;

[0020] Represents the maximum row sum of the total influence of all influencing factors, which is used for normalization to ensure that the value of the normalized direct influence matrix is ​​within an appropriate range;

[0021] Z3, according to the calculated normalized matrix B, according to the following formula:

[0022] Calculate the total impact matrix C;

[0023] Where C represents the total impact matrix, which includes the direct and indirect impacts of all factors in the system;

[0024] C ij Indicates that in the total influence matrix C, the total influence of factor i on factor j includes the accumulation of all direct and indirect influences of factor i on factor j;

[0025] B represents the normalized direct influence matrix, which is obtained by normalizing the initial direct influence matrix A and is used to represent the direct influence between various factors;

[0026] B 1 ,,,,B k It represents the power of the normalized matrix B, which represents the accumulation of indirect effects between factors as the transmission paths increase;

[0027] I represents the identity matrix, which is a matrix with diagonal influencing factors of 1 and other influencing factors of 0;

[0028] (IB) -1 represents the inverse matrix of matrix IB. According to the matrix geometric series, (IB) -1 It is used to represent the sum of all powers of B, thus simplifying the summation process of the total influence matrix;

[0029] Z4. Calculation of influence, influence, centrality and cause;

[0030] On the basis of obtaining the total impact matrix C, according to the formula:

[0031]

[0032] M i =D i +G i ;

[0033] R i =D i -G i ;

[0034] Calculate the influence degree Di, the influence degree Gi, the central degree Mi, and the cause degree Ri of the factors affecting financing guarantee in turn;

[0035] Among them, Di is the sum of the influencing factors of each row in the total influencing matrix C, which represents the sum of the influence of the corresponding factors of each row on other factors, that is, the degree of influence;

[0036] Gi is the sum of the influencing factors in each column of the total influencing matrix C, which means the sum of the influence of the corresponding factors in each column on other factors, that is, the degree of influence;

[0037] The centrality degree Mi is obtained by adding up the influence degree of each factor, which indicates the importance of the factor in the influencing factor index system. The larger the value, the greater the importance of the factor.

[0038] The cause degree Ri is composed of the difference between the influence degree and the affected degree. If the cause degree is greater than zero, it means that the factor has a greater impact on other factors and is called a cause factor.

[0039] On the contrary, factors that are more affected by other factors are called result factors, and a cause-effect relationship diagram is drawn with the centrality value (Mi) as the horizontal axis and the cause value (Ri) as the vertical axis.

[0040] A further technical improvement of the present invention is that in S3, the risk factors are arranged in sequence according to their levels, and the variables of each factor are connected in combination with the binary relationship between the risk factors to form a multi-layer hierarchical directed graph.

[0041] A further technical improvement of the present invention is to bind the influence matrix in step S2 with the real-time data of the corresponding enterprise, and dynamically adjust the influence matrix according to the changes in the real-time data to achieve dynamic risk assessment, which includes the following steps:

[0042] Q1. Data source definition;

[0043] Determine the specific data sources for each risk factor in the primary and secondary risk indicators;

[0044] Q2. Construct a dynamic direct impact matrix;

[0045] Regularly extract the latest risk data from the corresponding enterprise database, analyze the mutual influence between various risk factors by setting rules or machine learning models, and dynamically update the direct impact matrix A;

[0046] Normalize the changes in real-time data to obtain the normalized direct impact matrix B, and dynamically adjust the direct impact matrix B according to the trend of risk data changes;

[0047] Q3. Generation of dynamic multi-layer ladder directed graphs;

[0048] Based on the updated direct impact matrix B, recalculate the total impact matrix C;

[0049] Use a multi-layer hierarchical directed graph to display the hierarchical structure and transmission path of risk factors, and dynamically adjust the structure of the multi-layer hierarchical directed graph based on the latest impact relationship;

[0050] If the impact change of a risk factor exceeds the preset impact change threshold, the relationship paths and weights between related factors will be automatically adjusted in the multi-layer hierarchical directed graph to reflect the latest risk structure;

[0051] Q4. Dynamic trend analysis combined with historical data;

[0052] Record the historical impact data of each risk factor and use the time series analysis model to predict the future change trend of risk factors;

[0053] Combine historical data and forecast results to make advance adjustments to future changes that directly affect Matrix A and generate predictive risk transmission paths to help companies identify potential risks in advance;

[0054] Q5. Generate a risk assessment report in real time based on the dynamically adjusted hierarchical directed graph and the total impact matrix C.

[0055] A further technical improvement of the present invention is that in step Q3, the method for adjusting the relationship path and the weight includes:

[0056] Y1, real-time monitoring of changes in the direct impact matrix B;

[0057] After the data is updated, the change of each influencing factor in the direct influencing matrix B is checked. If it exceeds the preset influencing factor change threshold, it means that the direct influence degree of the corresponding two groups of influencing factors has changed, and based on the change of the corresponding influencing factors, a new relationship path is determined;

[0058] Y2, weight adjustment;

[0059] After the relationship path is determined, the impact strength of each path needs to be expressed by the edge weight, and it is ensured that the edge weights in the multi-layer hierarchical directed graph can dynamically reflect the latest impact relationship between risk factors, including:

[0060] y1. Recalculate the total impact matrix C based on the latest direct impact matrix B;

[0061] y2, edge weight adjustment;

[0062] In the multi-layer hierarchical directed graph, the weight of each edge is represented by the corresponding impact element in the total impact matrix to reflect the transmission intensity between risk factors;

[0063] y3. Establish a dynamic weight update mechanism;

[0064] If the value of a certain risk factor increases in the updated total impact matrix C, the weight of the edge in the multi-layer hierarchical directed graph increases accordingly, indicating that the impact of the corresponding risk factor has intensified;

[0065] y4, generate trigger conditions for edge weight update;

[0066] A trigger threshold ΔP is set to determine whether the edge weight needs to be adjusted. When the change of a risk factor in the total impact matrix C exceeds the trigger threshold ΔP, the update of the edge weight is triggered.

[0067] Compared with the prior art, the present invention has the following beneficial effects:

[0068] When constructing the risk indicator system, the present invention starts from the risk analysis of technology-based enterprise financing guarantee, incorporates risk factors such as technology-based enterprise financing guarantee into the risk indicator system, and also introduces the evaluation results of authoritative experts in the field of enterprise financing guarantee, so that the selection of risk factors is more scientific, reasonable and effective, thereby constructing a more scientific technology-based enterprise financing guarantee risk indicator evaluation system;

[0069] In addition, a multi-layer recursive directed graph of the risk factors of financing guarantees for technology-based enterprises is constructed, which can not only reflect the mutual influence of the risk factors of financing guarantees for technology-based enterprises, but also obtain the structure and process of the transmission between these risk factors, so as to quickly analyze the various risk factors and their correlations;

[0070] On the other hand, the present invention can also combine the DEMATEL-ISM method with real-time data. This solution realizes the dynamic assessment of the financing guarantee risk of technology-based enterprises. The direct impact matrix of each risk factor can be automatically updated according to the changes in real-time data, so that the risk assessment results can quickly respond to changes in the external environment and internal enterprise data, and provide enterprise managers with intuitive risk transfer paths and key nodes, which facilitates the rapid identification and response to major risks. BRIEF DESCRIPTION OF THE DRAWINGS

[0071] In order to facilitate understanding by those skilled in the art, the present invention is further described below with reference to the accompanying drawings.

[0072] Figure 1 This is a flow chart of Embodiment 1 of the present invention;

[0073] Figure 2 This is a multi-layer hierarchical directed graph according to Embodiment 1 of the present invention. DETAILED DESCRIPTION

[0074] In order to further explain the technical means and effects adopted by the present invention to achieve the predetermined invention purpose, the specific implementation methods, structures, features and effects of the present invention are described in detail below in conjunction with the accompanying drawings and preferred embodiments.

[0075] Example 1

[0076] See also Figure 1-2 As shown, the present invention provides a risk assessment method for technology-based enterprise financing guarantee based on DEMATEL-ISM, the method comprising the following steps:

[0077] S1. Based on the characteristics of the corresponding enterprise and the factors affecting its financing guarantee, obtain the influencing indicators of the security risk of the financing guarantee, establish a financing guarantee risk evaluation indicator system from the four dimensions of the corresponding enterprise's internal risk, guarantee business risk, macroeconomic risk, and industry market risk, and take the four types of risks as the first-level risk indicators, and the remaining influencing factors under the four types of risks as the second-level risk indicators, and label each first-level risk indicator and each second-level risk indicator;

[0078] The internal risks of an enterprise include core technology competitiveness risk M11, management team quality risk M12, and financial status risk M13;

[0079] Guarantee business risks include guarantee amount and term risk M21, counter-guarantee measures risk M22, and external emergency risk M23;

[0080] Macroeconomic risks include economic cycle fluctuation risk M31, policy environment change risk M32, and international environment uncertainty risk M33;

[0081] Industry market risks include market demand fluctuation risk M41, technology update and iteration risk M42, and supply chain management risk M43;

[0082] S2. Use DEMATEL to transform the interdependence of each influencing factor in the primary risk index and the secondary risk index into two causal groups, and quantitatively analyze the influence degree between the system elements. By calculating the influence degree matrix between each influencing factor, an influence relationship diagram is obtained, that is, the mutual influence between each wind direction influencing factor is transformed into nodes and edges in the influence relationship diagram, and the influence relationship diagram is used to find the key influencing factors;

[0083] The implementation steps include:

[0084] Z1. Directly affects the establishment of the relationship matrix;

[0085] Determine the set of influencing factors and assign scores to the degree of influence of each influencing factor on each other. The assigned scores represent the degree of influence of influencing factor Si on influencing factor Sj.

[0086] In this embodiment, the scoring method uses 5 experts and scholars who have been engaged in financial risk management for more than 5 years to evaluate and score the degree of influence of each causal influencing factor on each other, and the scoring is based on a 0-4 scoring standard, where no influence is 0 points, a weak influence is 1 point, a medium influence is 2 points, a strong influence is 3 points, and an extremely strong influence is 4 points;

[0087] And to avoid the differences in subjective perception and personal knowledge, the evaluation results of multiple experts were summarized using the average method to obtain the initial direct impact matrix;

[0088] Z2. Establishment of comprehensive impact matrix;

[0089] In order to eliminate the influence of different dimensions, according to the formula Normalize the direct influence matrix to obtain a normalized direct influence matrix B;

[0090] Among them, b ij represents the normalized direct influence, that is, the influence of influencing factor i on influencing factor j after normalization, b ij Used in direct impact matrix B;

[0091] a ij represents the influencing factors in the initial direct influence matrix, specifically the degree of influence of influencing factor i on influencing factor j, which is scored by experts and scholars in this embodiment;

[0092] Indicates the total influence of influencing factor i on other influencing factors;

[0093] Represents the maximum row sum of the total influence of all influencing factors, which is used for normalization to ensure that the value of the normalized direct influence matrix is ​​within an appropriate range;

[0094] Z3, according to the calculated normalized matrix B, according to the following formula:

[0095] Calculate the total impact matrix C;

[0096] Where C represents the total impact matrix, which includes the direct and indirect impacts of all factors in the system;

[0097] C ij Indicates the total influence of factor i on factor j in the total influence matrix C, including the accumulation of all direct and indirect influences of factor i on factor j;

[0098] B represents the normalized direct influence matrix, which is obtained by normalizing the initial direct influence matrix A and is used to represent the direct influence between various factors;

[0099] B 1 ,,,,B k It represents the power of the normalized matrix B, which represents the accumulation of indirect effects between factors as the transmission paths increase;

[0100] I represents the identity matrix, which is a matrix with diagonal influencing factors of 1 and other influencing factors of 0;

[0101] (IB) -1 represents the inverse matrix of matrix IB. According to the matrix geometric series, (IB) -1 It is used to represent the sum of all powers of B, thus simplifying the summation process of the total influence matrix;

[0102] Z4. Calculation of influence, influence, centrality and cause;

[0103] On the basis of obtaining the total impact matrix C, according to the formula:

[0104]

[0105] M i =D i +G i ;

[0106] R i =D i -G i ;

[0107] Calculate the influence degree Di, the influence degree Gi, the central degree Mi, and the cause degree Ri of the factors affecting financing guarantee in turn;

[0108] Among them, Di is the sum of the influencing factors of each row in the total influencing matrix C, which represents the sum of the influence of the corresponding factors of each row on other factors, that is, the degree of influence;

[0109] Gi is the sum of the influencing factors in each column of the total influencing matrix C, which means the sum of the influence of the corresponding factors in each column on other factors, that is, the degree of influence;

[0110] The centrality degree Mi is obtained by adding up the influence degree of each factor, which indicates the importance of the factor in the influencing factor index system. The larger the value, the greater the importance of the factor.

[0111] The cause degree Ri is composed of the difference between the influence degree and the affected degree. If the cause degree is greater than zero, it means that the factor has a greater impact on other factors and is called a cause factor.

[0112] On the contrary, it is more affected by other factors and is called the result factor. Draw a cause-effect relationship diagram with the centrality value (Mi) as the horizontal axis and the cause value (Ri) as the vertical axis;

[0113] The risk factors calculated by the DEMATEL-ISM model are divided into the following levels:

[0114] The risk factors included in the first layer are M21, M22, M42, and M43;

[0115] The risk factors included in the second layer are M11, M12, M13, M31, and M41;

[0116] The risk factors included in the third level are M32 and M33;

[0117] The risk factors included in the fourth level are M23;

[0118] At the same time, ISM is used to classify the influencing factors in the system and build a bottom-up multi-level hierarchical model, so as to more clearly present the hierarchical relationship between the influencing factors;

[0119] S3. Utilize directed graph analysis and visualization of multi-level hierarchical structure models to obtain multi-layer hierarchical directed graphs, and based on the multi-layer hierarchical directed graphs, obtain a multi-layer hierarchical explanatory structure model of the influencing factors of corporate financing guarantees, and obtain the influencing mechanism of each influencing factor on financing guarantees and the transmission relationship of each influencing factor from the model;

[0120] The financing guarantee safety network is constructed according to the multi-layer hierarchical explanatory structure model. When consensus is reached among the nodes of each influencing factor, consensus is reached only between the nodes that are mutually related and have a transmission relationship with the upper and lower layers.

[0121] Consensus is the unified understanding and adjustment of the transmission relationship between risk factors, which dynamically updates the status and weight of influencing factors by transmitting and feeding back information through interrelated nodes.

[0122] The consensus includes:

[0123] Interaction between upper and lower layers: When a risk factor changes, only the related nodes of the upper and lower layers will be used for information transmission and impact analysis, rather than global adjustments. This method reduces the participation of irrelevant nodes and optimizes computing efficiency;

[0124] Dynamic relationship adjustment: the consensus result will affect the relationship path and weight between upper and lower level nodes;

[0125] The goal of consensus is to ensure the dynamic consistency of the risk transmission network, so that the adjustment of each node can reflect the latest transmission relationship, reduce unnecessary global updates, and reach consensus only between relevant nodes at the upper and lower levels to ensure the efficiency and accuracy of the model;

[0126] Arrange the risk factors in order according to their levels, and connect the variables of each factor based on the binary relationship between the risk factors to form the following Figure 2 The multi-layer hierarchical directed graph shown.

[0127] Example 2

[0128] Compared with Example 1, Example 2 binds the influence matrix in step S2 with the real-time data of the corresponding enterprise, and dynamically adjusts the influence matrix according to the changes in the real-time data to achieve dynamic risk assessment, which includes the following steps:

[0129] Q1. Data source definition;

[0130] Determine the specific data sources for each risk factor in the primary and secondary risk indicators;

[0131] In this embodiment, the enterprise's inherent risk is bound to the financial data and technical data of the corresponding enterprise;

[0132] Guarantee business risk is bound to corresponding external event data and guarantee contract data;

[0133] Macroeconomic risks are tied to external economic indicator data and policy data;

[0134] Industry market risks are bound to market demand data and industry trend data;

[0135] The data source of each risk factor is updated in real time through the data interface, obtaining real-time data corresponding to the internal and external data of the enterprise;

[0136] Q2. Construct a dynamic direct impact matrix;

[0137] Regularly extract the latest risk data from the corresponding enterprise database, analyze the mutual influence between various risk factors by setting rules or machine learning models, and dynamically update the direct impact matrix A;

[0138] Normalize the changes in real-time data to obtain the normalized direct impact matrix B, and dynamically adjust it according to the data change trend;

[0139] Q3. Generation of dynamic multi-layer ladder directed graphs;

[0140] Based on the updated direct impact matrix B, recalculate the total impact matrix C;

[0141] Use a multi-layer hierarchical directed graph to display the hierarchical structure and transmission path of risk factors, and dynamically adjust the structure of the multi-layer hierarchical directed graph based on the latest impact relationship;

[0142] If the impact change of a risk factor exceeds the preset impact change threshold, the relationship paths and weights between related factors will be automatically adjusted in the multi-layer hierarchical directed graph to reflect the latest risk structure;

[0143] The adjustment methods of relationship paths and weights include:

[0144] Y1, real-time monitoring of changes in the direct impact matrix B;

[0145] After the data is updated, check each influencing factor b in the direct impact matrix B ij If b ij If the influencing factor change threshold is exceeded, it means that the direct influence of influencing factor i on influencing factor j has changed;

[0146] If b ij From 0 to non-0, that is and This means that a direct influence relationship from influencing factor i to influencing factor j has been added, and a response path needs to be added in the multi-layer hierarchical directed graph;

[0147] If b ij From non-0 to 0, that is and It means that the direct influence relationship from influencing factor i to influencing factor j is deleted, and the corresponding path needs to be removed in the multi-layer hierarchical directed graph;

[0148] Y2, weight adjustment;

[0149] After the relationship path is determined, the impact strength of each path needs to be expressed by the edge weight, and it is ensured that the edge weights in the multi-layer hierarchical directed graph can dynamically reflect the latest impact relationship between risk factors, including:

[0150] y1. Recalculate the total impact matrix C based on the latest direct impact matrix B;

[0151] y2, edge weight adjustment;

[0152] In a multi-layer hierarchical directed graph, the weight w of each edge ij The corresponding influence element C in the total influence matrix ij To express it, to reflect the transmission intensity between risk factors;

[0153] The edge weight calculation formula is w ij =k×c ij ;

[0154] where w ij is the weight of the edge from risk factor i to risk factor j, indicating the intensity of impact transmission;

[0155] C ij Indicates the total influence of factor i on factor j in the total influence matrix C;

[0156] k is a scaling factor used to map matrix values ​​to edge weights in the graph in order to display the relative strength of influence in a multi-layer hierarchical directed graph;

[0157] y3. Establish a dynamic weight update mechanism;

[0158] If in the updated total impact matrix C, a risk factor C ij As the value of increases, the edge w in the multi-layer hierarchical directed graph ij The weight of increases accordingly, indicating that the impact of risk factor i on risk factor j intensifies;

[0159] If C ij The value of decreases, then the edge w in the multi-layer hierarchical directed graph ij The weight of is reduced accordingly, indicating that the impact of risk factor i on risk factor j is weakened;

[0160] y4, generate trigger conditions for edge weight update;

[0161] A trigger threshold ΔP is set to determine whether the weight of the edge needs to be adjusted. When the change of a risk factor in the total impact matrix C exceeds the trigger threshold ΔP, the update of the edge weight is triggered. The formula is expressed as:

[0162]

[0163] in, is the value of C in the total influence matrix after update;

[0164] is the value of C in the total influence matrix before updating;

[0165] ΔP is the preset trigger threshold, which is used to control the update frequency of edge weights;

[0166] Q4. Dynamic trend analysis combined with historical data;

[0167] Record the historical impact data of each risk factor and use the time series analysis model to predict the future change trend of risk factors;

[0168] Combining historical data and forecast results, make advance adjustments to future changes that directly affect Matrix A and generate predictive risk transmission paths to help companies identify potential risks in advance;

[0169] Q5. Generate a risk assessment report in real time based on the dynamically adjusted hierarchical directed graph and the total impact matrix C.

[0170] The above description is only a preferred embodiment of the present invention and does not limit the present invention in any form. Although the present invention has been disclosed as a preferred embodiment as above, it is not used to limit the present invention. Any technical personnel in this field can make some changes or modify the technical contents disclosed above into equivalent embodiments without departing from the scope of the technical solution of the present invention. However, any simple modification, equivalent change and modification made to the above embodiments according to the technical essence of the present invention without departing from the content of the technical solution of the present invention still fall within the scope of the technical solution of the present invention.

Claims

1. A risk assessment method for financing guarantee of technology-based enterprises based on DEMATEL-I SM, characterized in that: The method comprises the following steps: S1. Based on the characteristics of the corresponding enterprise and the factors affecting its financing guarantee, obtain the influencing indicators of the security risk of the financing guarantee, establish a financing guarantee risk evaluation indicator system from the four dimensions of the corresponding enterprise's internal risk, guarantee business risk, macroeconomic risk, and industry market risk, and take the four types of risks as the first-level risk indicators, and the remaining influencing factors under the four types of risks as the second-level risk indicators, and label each first-level risk indicator and each second-level risk indicator; S2. Use DEMATEL to transform the interdependence of each influencing factor in the primary risk index and the secondary risk index into two causal groups, and quantitatively analyze the influence between the system elements. By calculating the influence matrix between each influencing factor, the influence relationship diagram is obtained, and the key influencing factors are found out by using the influence relationship diagram; At the same time, ISM is used to classify the influencing factors in the system and build a bottom-up multi-level hierarchical model; S3. Utilize directed graph analysis and visualization of multi-level hierarchical structure models to obtain multi-layer hierarchical directed graphs, and based on the multi-layer hierarchical directed graphs, obtain a multi-layer hierarchical explanatory structure model of the influencing factors of corporate financing guarantees, and obtain the influencing mechanism of each influencing factor on financing guarantees and the transmission relationship of each influencing factor from the model; Construct a financing guarantee safety network according to a multi-layer hierarchical explanatory structural model.

2. According to a DEMATEL-ISM-based risk assessment method for technology-based enterprise financing guarantees, the method is characterized in that: When consensus is reached among the nodes of each influencing factor, consensus is reached only between the nodes that are mutually associated with the upper and lower layers and have a conductive relationship.

3. The risk assessment method for technology-based enterprise financing guarantee based on DEMATEL-ISM according to claim 1 is characterized in that: The S2 step includes the following implementation steps: Z1. Directly affects the establishment of the relationship matrix; Determine the set of influencing factors and assign scores to the degree of influence of each influencing factor on each other; Z2. Establishment of comprehensive impact matrix; According to the formula Normalize the direct influence matrix to obtain a normalized direct influence matrix B; Among them, b ij represents the normalized direct influence; a ij Represents the influencing factors in the initial direct impact matrix; Indicates the total influence of influencing factor i on other influencing factors; Represents the maximum row sum of the total influence of all influencing factors; Z3, according to the calculated normalized matrix B, according to the following formula: Calculate the total impact matrix C; Where C represents the total impact matrix, which includes the direct and indirect impacts of all factors in the system; C ij Indicates the total influence of factor i on factor j in the total influence matrix C; B represents the normalized direct influence matrix; B 1 ,,,,B k It represents the power of the normalized matrix B, which represents the accumulation of indirect effects between factors as the transmission paths increase; I represents the identity matrix, which is a matrix with diagonal influencing factors of 1 and other influencing factors of 0; (IB) -1 represents the inverse matrix of matrix IB; Z4. Calculation of influence, influence, centrality and causality.

4. A risk assessment method for technology-based enterprise financing guarantee based on DEMATEL-ISM according to claim 3, characterized in that: In Z4, based on the total influence matrix C, according to the formula: M i =D i +G i ; R i =D i -G i ; Calculate the influence degree Di, influence degree Gi, centrality degree Mi and cause degree Ri of the factors affecting financing guarantee in turn.

5. The risk assessment method for technology-based enterprise financing guarantee based on DEMATEL-I SM according to claim 1 is characterized in that: In S3, the risk factors are arranged in sequence according to their levels, and the variables of each factor are connected in combination with the binary relationship between the risk factors to form a multi-layer hierarchical directed graph.

6. The risk assessment method for technology-based enterprise financing guarantee based on DEMATEL-ISM according to claim 1 is characterized in that: Bind the influence matrix in step S2 to the real-time data of the corresponding enterprise, and dynamically adjust the influence matrix according to the changes in the real-time data, which includes the following steps: Q1. Data source definition; Determine the specific data sources for each risk factor in the primary and secondary risk indicators; Q2. Construct a dynamic direct impact matrix; Regularly extract the latest risk data from the corresponding enterprise database, analyze the mutual influence between various risk factors by setting rules or machine learning models, and dynamically update the direct impact matrix A; Normalize the changes in real-time data to obtain the normalized direct impact matrix B, and dynamically adjust the direct impact matrix B according to the trend of risk data changes; Q3. Generation of dynamic multi-layer ladder directed graphs; Based on the updated direct impact matrix B, recalculate the total impact matrix C; Use a multi-layer hierarchical directed graph to display the hierarchical structure and transmission path of risk factors, and dynamically adjust the structure of the multi-layer hierarchical directed graph based on the latest impact relationship; Q4. Dynamic trend analysis combined with historical data; Record the historical impact data of each risk factor and use the time series analysis model to predict the future change trend of risk factors; Combining historical data and forecast results, make advance adjustments to future changes that directly affect matrix A and generate predictive risk transmission paths; Q5. Generate a risk assessment report in real time based on the dynamically adjusted hierarchical directed graph and the total impact matrix C.

7. A risk assessment method for technology-based enterprise financing guarantee based on DEMATEL-I SM according to claim 6, characterized in that: In step Q3, if the impact change of a risk factor exceeds the preset impact change threshold, the relationship paths and weights between related factors are automatically adjusted in the multi-layer hierarchical directed graph to reflect the latest risk structure.

8. A risk assessment method for technology-based enterprise financing guarantee based on DEMATEL-I SM according to claim 7, characterized in that: In step Q3, the adjustment methods of relationship paths and weights include: Y1, real-time monitoring of changes in the direct impact matrix B; After the data is updated, the changes of each influencing factor in the direct influencing matrix B are checked, and based on the changes of the corresponding influencing factors, a new relationship path is determined; Y2, weight adjustment; After the relationship path is determined, the impact strength of each path is expressed by the edge weight, including: y1. Recalculate the total impact matrix C based on the latest direct impact matrix B; y2, edge weight adjustment; y3. Establish a dynamic weight update mechanism; In the updated total impact matrix C, if the value of a certain risk factor increases, the weight of the edge in the multi-layer hierarchical directed graph will increase accordingly; y4, generate trigger conditions for edge weight update; A trigger threshold ΔP is set to determine whether the edge weight needs to be adjusted. When the change of a risk factor in the total impact matrix C exceeds the trigger threshold ΔP, the update of the edge weight is triggered.

Images