Financial risk analysis early warning system and method

Abstract

The invention relates to a financial risk analysis early warning system and method. The system comprises an equity relation network construction module which is used for collecting and processing equity relation data of a local financial institution and constructing a network topology with the financial institution as a node and the equity relation as an edge; the high-management relation network construction module is used for collecting and processing high-management social relation data of a local financial institution and constructing a weighted network with high management as nodes and social relations as edges; the double-layer network fusion module is used for fusing the stock right relation network and the high-management social relation network to generate a double-layer complex network large graph; and the risk analysis and early warning module is used for analyzing a conduction path and a propagation rule of the financial risk in the double-layer network, and carrying out quantitative evaluation and early warning on the financial risk faced by the local financial institution. According to the scheme provided by the invention, the data of the equity relationship and the social relationship can be deeply fused, and the requirements of supervision departments on comprehensive and accurate monitoring and early warning of financial risks are effectively met.

Description

Technical Field

[0001] The present invention relates to the field of financial management technology, and in particular to a financial risk analysis and early warning system and method. Background Art

[0002] Local financial institutions occupy a crucial position in the financial system, and their stable operation is crucial to local economic development. With the increasing complexity and diversification of financial operations, the interconnectedness among local financial institutions has become increasingly close, making the transmission of financial risks complex and dynamic. Equity relations are a crucial link between local financial institutions, reflecting their ownership structure and the control they exercise over them. Furthermore, the social connections of senior executives play a crucial role in the operations and decision-making processes of financial institutions, influencing the path and scope of financial risk transmission.

[0003] However, traditional financial risk analysis methods often analyze the equity relationships and social relationships of local financial institutions' executives relatively independently. For example, they analyze only the distribution of equity and shareholding ratios to determine the potential risk contagion paths and impact between financial institutions, ignoring the role of executives' social relationships. Furthermore, the depth of exploration of equity relationships is limited, making it difficult to fully and accurately reveal the complex equity connections and potential risk transmission mechanisms between financial institutions. Furthermore, when analyzing executives' social relationships, they often limit themselves to collating information such as their personal backgrounds and career experiences and conducting preliminary correlation analysis, failing to deeply integrate this information with their equity relationships. This results in a less comprehensive and in-depth assessment of financial risk, making it difficult to meet the needs of local financial regulators for comprehensive, accurate, and real-time monitoring and early warning of financial risks.

[0004] Therefore, there is an urgent need to design a financial risk analysis and early warning system and method that can effectively and deeply integrate the data of equity relations and executives' social relations, construct a large double-layer complex network diagram of equity relations and executives' social relations, give full play to the synergistic effect of the two in financial risk analysis, comprehensively reflect the complex correlation between financial institutions, accurately analyze the transmission rules, key nodes, transmission speed and other key information of financial risks between different financial institutions, and accurately identify potential risk transmission paths and nodes, effectively meeting the needs of local financial regulatory authorities for comprehensive, accurate and real-time monitoring and early warning of financial risks. Summary of the Invention

[0005] In order to overcome the problems existing in related technologies, the present application provides a financial risk analysis and early warning system and method. The financial risk analysis and early warning system and method can effectively deeply integrate the data of equity relations and social relations of senior executives, construct a two-layer complex network map of equity relations and social relations of senior executives, give full play to the synergistic effect of the two in financial risk analysis, comprehensively reflect the complex correlation relationship between financial institutions, accurately analyze the transmission rules, key nodes and transmission speed of financial risks between different financial institutions and other key information, and accurately identify potential risk transmission paths and nodes, effectively meeting the needs of local financial regulatory authorities for comprehensive, accurate and real-time monitoring and early warning of financial risks.

[0006] The first aspect of this application is to provide a financial risk analysis and early warning system, comprising:

[0007] The equity relationship network construction module is used to collect and process the equity relationship data of local financial institutions. It then constructs a network topology based on the processed equity relationship data, with financial institutions as nodes and equity relationships as edges, and identifies key equity nodes and important equity association paths.

[0008] The executive relationship network construction module is used to collect and process the social relationship data of executives of local financial institutions. It then constructs a weighted network based on the processed social relationship data, with executives as nodes and social relationships as edges, and identifies executive nodes with high centrality and influence, as well as groups of executives with close social relationships.

[0009] A two-layer network fusion module is used to connect the equity relationship network and the executive social relationship network through the relationship edge between local financial institutions and executives, fuse them to generate a two-layer complex network graph, and visualize the two-layer complex network graph;

[0010] The risk analysis and early warning module is used to label risk factors based on the double-layer complex network graph, and associate the risk factors with the nodes and edges in the double-layer complex network graph, analyze the transmission path and propagation law of financial risks in the double-layer network, and establish a risk assessment indicator system to quantitatively assess and warn of the financial risks faced by local financial institutions.

[0011] In the preferred technical solution of this application, the equity relationship network construction module includes:

[0012] A data collection unit, configured to collect equity relationship data of local financial institutions, wherein the equity relationship data includes but is not limited to business registration information, shareholder registers, equity change records, and equity pledge information data;

[0013] A data processing unit, used to clean, deduplicate, and standardize equity relationship data and build an equity relationship data model;

[0014] The network construction unit is used to draw a network topology diagram based on equity relationship data using network analysis software or programming language, and calculate the degree centrality and betweenness centrality of the nodes.

[0015] In the preferred technical solution of this application, the equity relationship network is assumed to be an undirected graph G = (V, E),

[0016] Where V represents the set of financial institution nodes, and E represents the set of equity relationship edges;

[0017] Node v i ∈V, its degree centrality C D (v i ) is calculated as:

[0018]

[0019] Among them, deg(v i ) represents the node v i The number of connected edges, i.e., node v i The degree of n is the total number of nodes in the equity relationship network.

[0020] In the preferred technical solution of this application, for node v i ∈V, its betweenness centrality C B (v i ) is calculated as:

[0021]

[0022] Among them, σ st represents the number of shortest paths from node s to node t, σ st (v i ) indicates passing through node v i The number of shortest paths.

[0023] In the preferred technical solution of this application, the executive relationship network construction module includes:

[0024] A data collection unit is used to collect social relationship data of executives, wherein the social relationship includes but is not limited to personal information, including name, position, professional resume, educational background, part-time social jobs, and personal network data;

[0025] The data processing unit is used to clean, classify, label and associate executive social relationship data, assign different weights to different types of relationships, and establish an executive social relationship data model;

[0026] The network construction unit is used to draw a network diagram based on the executives' social relationship data using network analysis software or programming language, analyze and determine the executive nodes with high centrality and influence, as well as the close social relationship groups of executives.

[0027] In a preferred technical solution of this application, the dual-layer network fusion module includes:

[0028] Node fusion unit: used to attach executive nodes to corresponding financial institution nodes based on the relationship between local financial institutions and executives, thus achieving the fusion of the two-layer network at the node level;

[0029] Edge fusion unit: used to assign different weights to each equity relationship and social relationship edge, and fuse the equity relationship edges and executive social relationship edges through weighted fusion to generate a two-layer complex network graph;

[0030] Visualization processing unit: Use visualization tools to visualize large graphs of two-layer complex networks.

[0031] In the preferred technical solution of this application, each equity relationship and social relationship edge is assigned a different weight, including:

[0032] Assume that the complex network graph after fusion is graph K = (V∪U, E∪F∪L),

[0033] Among them, V is the node set of financial institutions, U is the node set of executives, E is the edge set of equity relations, F is the edge set of executives' social relations, and L is the edge set of affiliation relations between financial institutions and executives;

[0034] When the side ab ∈E, its weight Among them, p is the weight coefficient of the equity relationship edge;

[0035] When the side f cd ∈F, its weight Among them, q is the weight coefficient of the executive’s social relationship edge, w cd is the edge weight of the executive’s social relationship;

[0036] When the side ef ∈L, its weight Where r is a constant;

[0037] And the coupling condition of p+q=1 is satisfied.

[0038] In a preferred technical solution of the present application, in the risk analysis and early warning module, the analysis of the transmission path and propagation rules of financial risks in the two-layer network includes:

[0039] The risk propagation probability model based on random walk analyzes the transmission path and propagation law of financial risks in a two-layer network. The calculation formula is:

[0040] Assume that the risk source node is s and the target node is d. In the complex network graph, the probability P of risk spreading from node s to node d is s→d It can be calculated by the following formula:

[0041]

[0042] Among them, N(s) represents the set of neighbor nodes of node s, w si represents the edge weight from node s to node i, α is the probability of random walk (usually around 0.85), δ sd is the indicator function. When s=d, δ sd =1, otherwise 0.

[0043] A second aspect of the present application is to provide a financial risk analysis and early warning method, which is implemented based on the above-mentioned financial risk analysis and early warning system and includes the following steps:

[0044] Construct an equity relationship network based on equity relationship data of local financial institutions;

[0045] Constructing a social relationship network of executives based on the social relationship data of executives of local financial institutions;

[0046] Merge the equity relationship network with the executive social relationship network to construct a double-layer complex network graph;

[0047] Based on the two-layer complex network diagram, local financial risks are analyzed and early warnings are issued.

[0048] In the preferred technical solution of this application, the following steps are specifically included:

[0049] Collect and process equity relationship data of local financial institutions, then build a network topology based on the processed equity relationship data, with financial institutions as nodes and equity relationships as edges, and identify key equity nodes and important equity association paths;

[0050] Collect and process the social relationship data of executives of local financial institutions. Then, based on the processed social relationship data, construct a weighted network with executives as nodes and social relationships as edges. This network then identifies executive nodes with high centrality and influence, as well as groups of executives with close social relationships.

[0051] Connect the relationship edges between local financial institutions and executives to the equity relationship network and the executives' social relationship network, fuse them to generate a double-layer complex network graph, and visualize the double-layer complex network graph;

[0052] The risk factors of the two-layer complex network graph are marked, and the risk factors are associated with the nodes and edges in the two-layer complex network graph, the transmission paths and propagation patterns of financial risks in the two-layer network are analyzed, and a risk assessment indicator system is established to conduct quantitative assessment and early warning of the financial risks faced by local financial institutions.

[0053] The technical solution provided by this application may include the following beneficial effects: The financial risk analysis and early warning system provided by the embodiment of this application can deeply integrate the correlation data of financial institution relationships and executive social relationships, give full play to the synergy between the two in financial risk analysis, and comprehensively and systematically reflect the complex correlation relationships between local financial institutions, breaking through the limitations of traditional single network analysis, and can accurately identify potential risk transmission paths, nodes and propagation laws, effectively improving the accuracy and timeliness of financial risk early warnings, helping regulatory authorities to better grasp the evolution trend and impact range of financial risks, so as to take timely measures to prevent and resolve financial risks. Moreover, by visually displaying a large double-layer complex network diagram, regulators and decision makers can more clearly understand the correlation relationships and risk status of local financial institutions, facilitate the formulation of more scientific and effective financial regulatory policies and risk prevention strategies, and improve financial regulatory efficiency and decision-making quality.

[0054] It should be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the present application. BRIEF DESCRIPTION OF THE DRAWINGS

[0055] The above and other objects, features and advantages of the present application will become more apparent through a more detailed description of exemplary embodiments of the present application in conjunction with the accompanying drawings, wherein the same reference numerals generally represent the same components in the exemplary embodiments of the present application.

[0056] Figure 1 It is a flow chart of the financial risk analysis and early warning method shown in an embodiment of the present application. DETAILED DESCRIPTION

[0057] The preferred embodiments of the present application will be described in more detail below with reference to the accompanying drawings. Although the preferred embodiments of the present application are shown in the accompanying drawings, it should be understood that the present application can be implemented in various forms and should not be limited by the embodiments described herein. Instead, these embodiments are provided to make the present application more thorough and complete, and to fully convey the scope of the present application to those skilled in the art.

[0058] In traditional financial risk analysis methods, the analysis of equity relations and social relations of senior executives of local financial institutions is mostly relatively independent, and the two are not deeply integrated, resulting in an insufficiently comprehensive and in-depth assessment of financial risks, making it difficult to meet the needs of local financial regulatory authorities for comprehensive, accurate, and real-time monitoring and early warning of financial risks.

[0059] In response to the above problems, the embodiments of the present application provide a financial risk analysis and early warning system and method, which can effectively and deeply integrate the data of equity relations and social relations of senior executives, construct a two-layer complex network diagram of equity relations and social relations of senior executives, give full play to the synergistic effect of the two in financial risk analysis, comprehensively reflect the complex correlation relationship between financial institutions, accurately analyze the transmission rules, key nodes and transmission speed of financial risks between different financial institutions and other key information, and accurately identify potential risk transmission paths and nodes, effectively meeting the needs of local financial regulatory authorities for comprehensive, accurate and real-time monitoring and early warning of financial risks.

[0060] The technical solutions of the embodiments of the present application are described in detail below with reference to the accompanying drawings.

[0061] Example 1

[0062] The financial risk analysis and early warning system of the present application includes: an equity relationship network construction module, an executive relationship network construction module, a two-layer network fusion module and a risk analysis and early warning module. The equity relationship network construction module is used to collect and process the equity relationship data of local financial institutions, and then construct a network topology with financial institutions as nodes and equity relationships as edges based on the processed equity relationship data, and determine key equity nodes and important equity association paths; the executive relationship network construction module is used to collect and process the social relationship data of executives of local financial institutions, and then construct a weighted network with executives as nodes and social relationships as edges based on the processed social relationship data, and determine the network with higher centrality. and influential executive nodes, as well as close social relationship groups of executives; the two-layer network fusion module is used to connect the equity relationship network and the executive social relationship network through the local financial institutions and executive relationship edges, fuse to generate a two-layer complex network large graph, and visualize the two-layer complex network large graph; the risk analysis and early warning module is used to label risk factors based on the two-layer complex network large graph, and associate the risk factors with the nodes and edges in the two-layer complex network large graph, analyze the transmission path and propagation law of financial risks in the two-layer network, and establish a risk assessment indicator system to quantitatively assess and warn of the financial risks faced by local financial institutions.

[0063] Specifically, the equity relationship network construction module includes: a data collection unit, a data processing unit and a network construction unit. The data collection unit is used to collect equity relationship data of local financial institutions, wherein the equity relationship data includes but is not limited to industrial and commercial registration information, shareholder registers, equity change records and equity pledge information data. In actual applications, the equity relationship data can be obtained from local financial regulatory departments, industrial and commercial administrative departments, information disclosed by the financial institutions themselves, and professional financial data service agencies, so as to ensure the comprehensiveness and accuracy of the data.

[0064] The data processing unit is used to clean, deduplicate, and standardize equity relationship data and construct an equity relationship data model. Specifically, the collected equity relationship data is cleaned, deduplicated, and standardized, the data format is unified, and issues such as missing values, erroneous values, and duplicate records are resolved. Furthermore, the relationships between key attributes and entities, such as financial institutions, shareholders, shareholding ratios, and equity change dates, are clarified.

[0065] The network construction unit is used to draw a network topology diagram based on the equity relationship data using network analysis software or programming language, and calculate the degree centrality and betweenness centrality of the nodes. Specifically, based on the processed equity relationship data, an equity relationship network can be constructed with local financial institutions as nodes and equity relationships as edges, and a network analysis software (such as Gephi, Pajek, etc.) or programming language (such as the NetworkX library in Python) can be used to draw a network topology diagram. Then, by calculating indicators such as the degree centrality and betweenness centrality of the nodes, the key equity nodes and important equity association paths in the network are determined, and the equity associations and control relationships between financial institutions are preliminarily revealed. For example, the equity relationship network is assumed to be an undirected graph G = (V, E), where V represents the set of financial institution nodes and E represents the set of equity relationship edges. For node v i ∈V, its degree centrality C D (v i ) is calculated as:

[0066]

[0067] Among them, deg(v i ) represents the node v i The number of connected edges, i.e., node v i The degree of n is the total number of nodes in the equity relationship network.

[0068] For node v i ∈V, its betweenness centrality C B (v i ) is calculated as:

[0069]

[0070] Among them, σ st represents the number of shortest paths from node s to node t, σ st (v i ) indicates passing through node v i The number of shortest paths.

[0071] By calculating degree centrality, we can understand the number of direct connections between financial institution nodes in the equity relationship network. D (v i ) value, the more direct equity connections the financial institution has with other financial institutions, and the higher its activity in the equity relationship network. By calculating the betweenness centrality, we can understand the control and influence of financial institution nodes in the equity network. If C B (v i ) value, the larger the value is, the more critical the financial institution node plays a bridge role in the dissemination of equity relations and information transmission, and is a key node for the transmission of financial risks.

[0072] The executive relationship network construction module includes: a data collection unit, a data processing unit and a network construction unit. The data collection unit is used to collect the social relationship data of executives, wherein the social relationships include but are not limited to personal information, including name, position, professional resume, educational background, social part-time jobs, and personal network relationship data. In actual applications, the social relationships can be obtained from public channels such as the official website of financial institutions, annual reports, news reports, industry forums, professional social platforms (such as LinkedIn), etc. At the same time, some non-public but important executive social relationship information can also be collected through field research, questionnaires, etc.

[0073] The data processing unit is used to clean, classify, label and associate the executives' social relationship data, and then assign different weights according to different types of relationships, and establish an executive social relationship data model. Specifically, the executives' social relationship data is cleaned, classified, labeled and associated to remove duplicate and irrelevant information, and the executives' social relationships are classified, such as classmates, colleagues, relatives, partners, etc., and then different weights are assigned according to different types of relationships to establish an executive social relationship data model, clarifying the relationship structure between attributes such as individual executives, social relationship types, and relationship strength. Exemplarily, the different weights assigned according to different types of relationships are specifically:

[0074] Suppose the executive social relationship network is a graph H = (U, F), where U represents the set of executive nodes and F represents the set of social relationship edges. jk ∈F, connecting executive u j and u k, whose weight w jk Determined by relationship type and strength:

[0075]

[0076] Among them, w1>w2>w3>w4>0, such as w1=0.9, w2=0.7, w3=0.6, w4=0.4.

[0077] By assigning different weights to the edges of social relationships, the closeness of social relationships between executives can be reflected. The larger the weight, the closer the social relationship between the two people, and the greater the impact it may have on the transmission of financial risks. For example, when w1>w2, it means that the relationship between relatives is closer than the relationship between classmates, and the impact it produces is greater.

[0078] The network construction unit is used to draw a network diagram based on the executive social relationship data using network analysis software or a programming language, analyze and identify executive nodes with high centrality and influence, as well as close executive social relationship groups. Specifically, the executive social relationship network is constructed with individual executives as nodes and social relationships as edges. The network diagram is drawn using network analysis software (such as Gephi, Pajek, etc.) or a programming language (such as the NetworkX library in Python), and indicators such as network density, average path length, and clustering coefficient are analyzed to identify executive nodes with high centrality and influence, as well as close executive social relationship groups, providing a basis for subsequent analysis of the impact of executive social relationships on the transmission of financial risks.

[0079] The two-layer network fusion module includes a node fusion unit, an edge fusion unit, and a visualization processing unit. The node fusion unit is used to attach the executive node to the corresponding financial institution node based on the relationship between the local financial institution and the executive, thereby achieving node-level fusion of the two-layer network. The edge fusion unit is used to assign different weights to each equity relationship and social relationship edge, and to fuse the equity relationship edges and the executive social relationship edges through weighted fusion to generate a large two-layer complex network diagram. The visualization processing unit uses visualization tools to visualize the large two-layer complex network diagram.

[0080] Specifically, the node fusion unit maps and associates the financial institution nodes in the equity relationship network with the executive nodes in the executive social relationship network. For example, based on the affiliation between the financial institution and the executive, the executive node is attached to the corresponding financial institution node, achieving node-level fusion of the two networks. Simultaneously, an edge is established between the financial institution node and the executive node, representing the affiliation and influence relationship between the financial institution and its executives.

[0081] The edge fusion unit is used to assign different weights to each equity relationship and social relationship edge. For example, the weight of the equity relationship edge can be determined according to the shareholding ratio. The higher the shareholding ratio, the greater the weight. The weight of the executive social relationship edge is determined according to the relationship type and strength. For example, the weight of the kinship relationship is higher than that of the ordinary classmate relationship. Then, through weighted fusion, the edges of the two-layer network are unified in a complex network graph, constructing a two-layer complex network graph that can comprehensively reflect the equity and executive social relations of local financial institutions.

[0082] Specifically, each equity relationship and social relationship edge is assigned a different weight, including:

[0083] Assume that the complex network graph after fusion is graph K = (V∪U, E∪F∪L),

[0084] Among them, V is the node set of financial institutions, U is the node set of executives, E is the edge set of equity relations, F is the edge set of executives' social relations, and L is the edge set of affiliation relations between financial institutions and executives;

[0085] When the side ab ∈E, its weight Among them, p is the weight coefficient of the equity relationship edge;

[0086] When the side f cd ∈F, its weight Among them, q is the weight coefficient of the executive’s social relationship edge, w cd is the edge weight of the executive’s social relationship;

[0087] When the side ef ∈L, its weight Where r is a constant;

[0088] And the coupling condition of p+q=1 is satisfied; it should be noted that (1) if the impact of equity relations on the transmission of financial risks is considered to be greater than that of executive social relations, then p>q can be set; (2) the value of r is between the minimum and maximum values ​​of the weights of the equity relationship edge and the executive social relationship edge.

[0089] The visualization processing unit uses professional graphic visualization tools (such as D3.js, ECharts, etc.) to visualize the constructed two-layer complex network diagram. For example, different graphic elements (such as node size, color, shape, edge thickness, color, etc.) can be used to intuitively display the position, importance and strength of the relationship between financial institutions and executives in the two-layer network, making it convenient for users to observe and analyze the complex network diagram, and providing strong support for the visualization analysis of financial risks.

[0090] The risk analysis and early warning module includes a risk factor labeling unit, a risk transmission path analysis unit and a risk assessment and early warning unit. The risk factor labeling unit is used to identify and label factors that may cause financial risks, such as abnormal financial indicators such as the non-performing loan rate, capital adequacy ratio, debt-to-asset ratio of financial institutions, equity risk events such as excessive equity pledges and frequent equity changes, and executive risk factors such as illegal and disciplinary behavior of executives and frequent changes. These risk factors are associated with nodes and edges in the network to provide risk source information for subsequent risk transmission analysis. The risk transmission path analysis unit is used to analyze the transmission path and propagation law of financial risks in a two-layer network based on a large double-layer complex network graph and using a random walk risk propagation model. Specifically, it includes: assuming that the risk source node is s and the target node is d, in the large double-layer complex network graph, the risk propagation probability P of risk propagating from node s to node d is s→d Calculated by the following formula:

[0091]

[0092] Among them, N(s) represents the set of neighbor nodes of node s, w si represents the edge weight from node s to node i, α is the probability of random walk (usually around 0.85), δ sd is the indicator function. When s=d, δ sd =1, otherwise 0.

[0093] By calculating the probability of risk transmission, we can determine the risk transmission path starting from the risk source node and gradually spreading to other financial institution nodes through equity relations and executive social relations, as well as the key transmission nodes and risk transmission intensity on each path, to help regulators and financial institutions identify and warn of the potential risk transmission scope and impact in advance.

[0094] The risk assessment and early warning unit is used to establish a risk assessment indicator system based on factors such as the number of risk factors of each node in the double-layer complex network diagram, the weight accumulation on the risk transmission path, the centrality of the node in the network, etc., to conduct quantitative assessment and early warning of the financial risks faced by local financial institutions, and based on the risk assessment results, propose corresponding regulatory measures and risk prevention suggestions for financial institutions of different risk levels to achieve effective management and control of local financial risks.

[0095] In the first embodiment of the present invention, the financial risk analysis and early warning system provided by the embodiment of the present application can deeply integrate the correlation data of financial institution relationships and executive social relationships, give full play to the synergistic effect of the two in financial risk analysis, and comprehensively and systematically reflect the complex correlation relationships between local financial institutions. It breaks through the limitations of traditional single network analysis and can accurately identify potential risk transmission paths, nodes and propagation laws, effectively improving the accuracy and timeliness of financial risk early warnings, helping regulatory authorities to better grasp the evolution trend and impact range of financial risks, so as to take timely measures to prevent and resolve financial risks. Moreover, by visually displaying a large double-layer complex network diagram, regulators and decision makers can more clearly understand the correlation relationships and risk status of local financial institutions, facilitate the formulation of more scientific and effective financial regulatory policies and risk prevention strategies, and improve financial regulatory efficiency and decision-making quality.

[0096] Example 2

[0097] Corresponding to the above-mentioned financial risk analysis and early warning system, this application also proposes a financial risk analysis and early warning method, please refer to Figure 1 , specifically:

[0098] Based on the structure of the above-mentioned embodiment 1, a financial risk analysis and early warning method provided in embodiment 2 of the present application is implemented based on the above-mentioned financial risk analysis and early warning system, and includes the following steps:

[0099] S1. Construct an equity relationship network based on equity relationship data of local financial institutions;

[0100] S2. Constructing a social relationship network of executives based on the social relationship data of executives of local financial institutions;

[0101] S3. Merge the equity relationship network with the executive social relationship network to construct a double-layer complex network graph;

[0102] S4. Based on the two-layer complex network diagram, analyze local financial risks and issue warnings.

[0103] Furthermore, in S1, the step of constructing an equity relationship network based on the equity relationship data of local financial institutions specifically includes the following steps:

[0104] Collect and process the equity relationship data of local financial institutions, and then build a network topology with financial institutions as nodes and equity relationships as edges based on the processed equity relationship data, and determine the key equity nodes and important equity association paths.

[0105] In S2, the construction of the executive social relationship network based on the social relationship data of executives of local financial institutions specifically includes the following steps:

[0106] Collect and process the social relationship data of executives of local financial institutions. Then, based on the processed social relationship data, construct a weighted network with executives as nodes and social relationships as edges. This network then identifies executive nodes with high centrality and influence, as well as groups of executives with close social relationships.

[0107] In S3, the equity relationship network and the executive social relationship network are integrated to construct a two-layer complex network graph, which specifically includes the following steps:

[0108] Connect the relationship edges between local financial institutions and executives to the equity relationship network and the executives' social relationship network, fuse them to generate a double-layer complex network graph, and visualize the double-layer complex network graph;

[0109] In S4, the analysis of local financial risks and early warning based on the two-layer complex network diagram specifically includes the following steps:

[0110] The risk factors of the two-layer complex network graph are marked, and the risk factors are associated with the nodes and edges in the two-layer complex network graph, the transmission paths and propagation patterns of financial risks in the two-layer network are analyzed, and a risk assessment indicator system is established to conduct quantitative assessment and early warning of the financial risks faced by local financial institutions.

[0111] Those skilled in the art will further appreciate that the various illustrative logical blocks, modules, circuits, and algorithm steps described in connection with the application herein may be implemented as electronic hardware, computer software, or combinations of both.

[0112] The flowcharts and block diagrams in the accompanying drawings illustrate possible architectures, functions, and operations of systems and methods according to various embodiments of the present application. In some alternative implementations, the functions marked in the blocks may occur in an order different from that marked in the accompanying drawings. For example, two consecutive blocks may actually be executed substantially in parallel, or they may sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each block in the block diagrams and / or flow charts, and combinations of blocks in the block diagrams and / or flow charts, may be implemented using a dedicated hardware-based system that performs the specified functions or operations, or may be implemented using a combination of dedicated hardware and computer instructions.

[0113] The embodiments of the present application have been described above. The above description is illustrative and not exhaustive, and is not limited to the disclosed embodiments. Many modifications and variations will be apparent to those skilled in the art without departing from the scope and spirit of the described embodiments. The terminology used herein is selected to best explain the principles of the embodiments, their practical applications, or improvements to the technology in the market, or to enable other persons skilled in the art to understand the embodiments disclosed herein.

Claims

1. A financial risk analysis and early warning system, characterized in that: include: The equity relationship network construction module is used to collect and process the equity relationship data of local financial institutions. It then constructs a network topology based on the processed equity relationship data, with financial institutions as nodes and equity relationships as edges, and identifies key equity nodes and important equity association paths. The executive relationship network construction module is used to collect and process the social relationship data of executives of local financial institutions. It then constructs a weighted network based on the processed social relationship data, with executives as nodes and social relationships as edges, and identifies executive nodes with high centrality and influence, as well as groups of executives with close social relationships. A two-layer network fusion module is used to connect the equity relationship network and the executive social relationship network through the relationship edge between local financial institutions and executives, fuse them to generate a two-layer complex network graph, and visualize the two-layer complex network graph; The risk analysis and early warning module is used to label risk factors based on the double-layer complex network graph, and associate the risk factors with the nodes and edges in the double-layer complex network graph, analyze the transmission path and propagation law of financial risks in the double-layer network, and establish a risk assessment indicator system to quantitatively assess and warn of the financial risks faced by local financial institutions.

2. The financial risk analysis and early warning system according to claim 1, characterized in that: The equity relationship network construction module includes: A data collection unit, configured to collect equity relationship data of local financial institutions, wherein the equity relationship data includes but is not limited to business registration information, shareholder registers, equity change records, and equity pledge information data; A data processing unit, used to clean, deduplicate, and standardize equity relationship data and build an equity relationship data model; The network construction unit is used to draw a network topology diagram based on equity relationship data using network analysis software or programming language, and calculate the degree centrality and betweenness centrality of the nodes.

3. The financial risk analysis and early warning system according to claim 2, characterized in that: Assume that the equity relationship network is an undirected graph G = (V, E), Where V represents the set of financial institution nodes, and E represents the set of equity relationship edges; Node v i ∈V, its degree centrality C D (v i ) is calculated as: Among them, deg(v i ) represents the node v i The number of connected edges, i.e., node v i The degree of n is the total number of nodes in the equity relationship network.

4. The financial risk analysis and early warning system according to claim 3, characterized in that: For node v i ∈V, its betweenness centrality C B (v i ) is calculated as: Among them, σ st represents the number of shortest paths from node s to node t, σ st (v i ) indicates passing through node v i The number of shortest paths.

5. The financial risk analysis and early warning system according to claim 1, characterized in that: The executive relationship network building module includes: A data collection unit is used to collect social relationship data of executives, wherein the social relationship includes but is not limited to personal information, including name, position, professional resume, educational background, part-time social jobs, and personal network data; The data processing unit is used to clean, classify, label, and associate executive social relationship data, assign different weights to different types of relationships, and establish an executive social relationship data model; The network construction unit is used to draw a network diagram based on the executives' social relationship data using network analysis software or programming language, analyze and determine the executive nodes with high centrality and influence, as well as the close social relationship groups of executives.

6. The financial risk analysis and early warning system according to claim 1, characterized in that: The dual-layer network fusion module includes: Node fusion unit: used to attach executive nodes to corresponding financial institution nodes based on the relationship between local financial institutions and executives, thus achieving the fusion of the two-layer network at the node level; Edge fusion unit: used to assign different weights to each equity relationship and social relationship edge, and fuse the equity relationship edges and executive social relationship edges through weighted fusion to generate a two-layer complex network graph; Visualization processing unit: Use visualization tools to visualize large graphs of two-layer complex networks.

7. The financial risk analysis and early warning system according to claim 6, characterized in that: Each equity relationship and social relationship edge is assigned a different weight, including: Assume that the complex network graph after fusion is graph K = (V∪U, E∪F∪L), Among them, V is the node set of financial institutions, U is the node set of executives, E is the edge set of equity relations, F is the edge set of executives' social relations, and L is the edge set of affiliation relations between financial institutions and executives; When the side ab ∈E, its weight Among them, p is the weight coefficient of the equity relationship edge; When the side f cd ∈F, its weight Among them, q is the weight coefficient of the executive’s social relationship edge, w cd is the edge weight of the executive’s social relationship; When the side ef ∈L, its weight Where r is a constant; And the coupling condition of p+q=1 is satisfied.

8. The financial risk analysis and early warning system according to claim 1, characterized in that: In the risk analysis and early warning module, the analysis of the transmission path and propagation rules of financial risks in the two-layer network includes: The risk propagation probability model based on random walk analyzes the transmission path and propagation law of financial risks in a two-layer network. The calculation formula is: Assume that the risk source node is s and the target node is d. In the complex network graph, the probability P of risk spreading from node s to node d is s→d It can be calculated by the following formula: Among them, N(s) represents the set of neighbor nodes of node s, w si represents the edge weight from node s to node i, α is the probability of random walk (usually around 0.85), δ sd is the indicator function. When s=d, δ sd =1, otherwise 0.

9. A financial risk analysis and early warning method, characterized in that: The implementation of the financial risk analysis and early warning system according to any one of claims 1 to 8 includes the following steps: Construct an equity relationship network based on equity relationship data of local financial institutions; Constructing a social relationship network of executives based on the social relationship data of executives of local financial institutions; Merge the equity relationship network with the executive social relationship network to construct a double-layer complex network graph; Based on the two-layer complex network diagram, local financial risks are analyzed and early warnings are issued.

10. The financial risk analysis and early warning method according to claim 9, characterized in that: The specific steps include: Collect and process equity relationship data of local financial institutions, then build a network topology based on the processed equity relationship data, with financial institutions as nodes and equity relationships as edges, and identify key equity nodes and equity association paths; Collect and process the social relationship data of executives of local financial institutions. Then, based on the processed social relationship data, construct a weighted network with executives as nodes and social relationships as edges. This network then identifies executive nodes with high centrality and influence, as well as groups of executives with close social relationships. Connect the relationship edges between local financial institutions and executives to the equity relationship network and the executives' social relationship network, fuse them to generate a double-layer complex network graph, and visualize the double-layer complex network graph; The risk factors of the two-layer complex network graph are marked, and the risk factors are associated with the nodes and edges in the two-layer complex network graph, the transmission paths and propagation patterns of financial risks in the two-layer network are analyzed, and a risk assessment indicator system is established to conduct quantitative assessment and early warning of the financial risks faced by local financial institutions.

Images