Smart city fine management platform based on GIS

Abstract

The invention, which relates to the technical field of smart city fine management, discloses a GIS-based smart city fine management platform comprising a GIS acquisition module, a multi-dimensional analysis conversion module, a multi-dimensional visualization module, a visualization analysis module and a visualization report module. In the prior art, an urban governance platform generally depends on single-dimensional data and is difficult to comprehensively reflect the complexity of urban operation; the deep correlation analysis is lacked, so that the relationship between different types of data is difficult to identify, and the decision accuracy is influenced; the method is limited to simple chart display, lacks interactivity and dynamics, and cannot effectively support real-time decision; according to the invention, through multi-dimensional data classification and correlation analysis, the complex relationship between urban data is comprehensively revealed; the dynamic visual network diagram with high interactivity is utilized, the real-time performance and accuracy of decision support are improved, the defects that in a traditional system, data display is simple, and real-time decision cannot be supported are overcome, and the efficiency and response capability of urban governance are enhanced.

Description

Technical Field

[0001] The present invention relates to the technical field of refined management of smart cities, and in particular to a refined management platform of smart cities based on GIS. Background Art

[0002] The GIS-based smart city refined management platform uses geographic information system (GIS) technology to integrate multi-source data of the city (such as transportation, environment, energy, public safety, etc.) to achieve real-time monitoring, analysis and intelligent scheduling of various resources and activities in the city. Through efficient data visualization, multi-dimensional data analysis, predictive modeling and intelligent decision support system, the platform can accurately locate problems, optimize resource allocation, and improve management efficiency, thereby promoting the transformation of urban governance from traditional experience-based management to data-driven intelligent management. This data-based decision-making model not only improves the transparency and efficiency of urban operations, but also enhances the ability to deal with complex and dynamic urban problems, such as traffic congestion management and environmental protection.

[0003] However, the platform also faces many challenges and defects in practical applications. First, existing urban governance platforms usually rely on single-dimensional data (such as geographic area or time dimension), which makes it difficult to fully reflect the complexity of urban operations. Second, existing systems often lack in-depth correlation analysis, which makes it difficult to identify the relationship between different types of data, affecting the accuracy of decision-making. Finally, data visualization in existing technologies is often limited to simple chart displays, lacks interactivity and dynamism, and cannot effectively support real-time decision-making. Summary of the invention

[0004] The purpose of the present invention is to solve the above-mentioned problems and to propose a GIS-based smart city refined management platform.

[0005] The purpose of the present invention can be achieved by the following technical solutions: a GIS-based smart city refined management platform, including a multidimensional analysis conversion module, a multidimensional visualization module, a visualization analysis module and a visualization report module; the multidimensional analysis conversion module is used to classify the urban data set to obtain the classified data, and then perform correlation analysis to obtain the Pearson correlation coefficient r, the Spearman rank correlation coefficient ρ and the square statistic between the two data ;

[0006] The visualization module is used to obtain various classified data for visualization;

[0007] The visualization analysis module is used to analyze the data after visualization connection, specifically:

[0008] Calculate the strength of each node to get the sum of the node edge weights ;

[0009] Calculate the centrality of the node in the network to get the centrality coefficient and betweenness centrality coefficient ;

[0010] By analyzing the number of paths to node i1, we can determine whether the relationship between the node and other nodes is balanced and obtain the node heterogeneity coefficient. ;

[0011] Then analyze according to the importance and heterogeneity of the nodes: calculate the deviation between the node strength and the mean strength of all nodes ; Then calculate whether the correlation between the node and other nodes is abnormal and output the abnormal coefficient of the relationship ; The calculated relationship anomaly coefficient , degree of deviation and betweenness centrality coefficient Comprehensive calculation to obtain the risk abnormality coefficient ;

[0012] The visual report module is used to obtain the risk anomaly coefficient Carry out refined management, specifically:

[0013] Risk Abnormality Coefficient Perform data hierarchical management: Compared with the preset threshold, if the risk abnormal coefficient <threshold T1, it is a low risk layer; if threshold T1<risk abnormality coefficient < threshold T2, it is a medium risk layer; if the risk abnormality coefficient ≥Threshold T2, it is a high-risk layer.

[0014] As a preferred embodiment of the present invention, it also includes a GIS acquisition module;

[0015] The GIS acquisition module is used to collect multi-source data of the city, including transportation, environment, energy and public safety, and integrate the multi-source data, combining the spatial data and non-spatial data in the geographic information system to obtain the city data set; and deduplicate, denoise and standardize the city data set to ensure the accuracy and consistency of the data.

[0016] As a preferred embodiment of the present invention, the specific process of performing correlation analysis to obtain the Pearson correlation coefficient r between two sets of data is:

[0017] The two corresponding data are calculated as data A and data B, and each data contains several variables. Each variable may represent the data of a time point or a region. The linear relationship between data A and data B is analyzed by the Pearson correlation coefficient: From the formula: Output the Pearson correlation coefficient r between x and y; where x and y are the variable values ​​in data A and data B respectively; n is the number of data points; ∑xy is the sum of the products of each pair of corresponding data in data A and data B; ∑x and ∑y are the sums of all data points in data A and data B respectively.

[0018] As a preferred embodiment of the present invention, the specific process of performing correlation analysis to obtain the Spearman rank correlation coefficient ρ between two sets of data is as follows:

[0019] If the data is ranked data, use the Spearman rank correlation coefficient to calculate: by the formula: Output the Spearman rank correlation coefficient ρ between x and y, d i is the ranking difference of the i-th pair of data, , n is the number of data points.

[0020] As a preferred embodiment of the present invention, a correlation analysis is performed to obtain the chi-square statistic between the two sets of data. The specific process is:

[0021] Two categorical data sets, calculated using the chi-square test: by the formula: Output the chi-square statistic between x and y , where O i is the observed frequency; E i is the expected frequency, and E i = (row sum) × (column sum) / total number of samples.

[0022] As a preferred embodiment of the present invention, the specific process of the visualization module for obtaining various classified data for visualization is as follows:

[0023] Obtain the classified data set and its correlation analysis results, and map the classified data into a network diagram for intuitive display; the classified data is presented in the form of nodes, and the color and size of the nodes reflect the characteristics of the data quantity or weight, while the positively correlated data nodes are connected by solid lines, and the color and thickness of the line indicate the strength of the correlation; the negatively correlated nodes are connected by dotted lines, and visual optimization is also performed based on the strength of the correlation; the data nodes without correlation remain independent and are not connected.

[0024] As a preferred embodiment of the present invention, the strength of each node is calculated to obtain the sum of the node edge weights. The specific process is:

[0025] Calculate the strength of each node, the sum of all edge weights of the node, by the formula: Output node edge weight sum , Represents the strength of the correlation between nodes i1 and j1.

[0026] As a preferred embodiment of the present invention, the centrality of the computing node in the network is obtained by: and betweenness centrality coefficient The specific process is:

[0027] Calculate the centrality of a node in the network and select different centrality measures. The centrality coefficient is: the number of edges directly connected to the node is ; Betweenness centrality coefficient: indicates the importance of a node in the shortest path, the formula is: Output betweenness centrality coefficient , is the number of shortest paths from node s1 to t1, is the number of paths passing through node i1.

[0028] As a preferred embodiment of the present invention, the risk abnormality coefficient obtained by comprehensive calculation is The specific process is:

[0029] Calculate the deviation between the node strength and the mean strength of all nodes using the formula: Output deviation ,in, is the mean value of all node strengths, is the standard deviation; then calculate whether the correlation between the node and other nodes is abnormal, through the formula: Output relation anomaly coefficient , is the mean of the edge weights between all nodes; the calculated relationship anomaly coefficient , degree of deviation and betweenness centrality coefficient Perform comprehensive calculations using the set formula: Output risk abnormality coefficient , where b1, b2 and b3 are preset weight parameters.

[0030] Compared with the prior art, the present invention has the following beneficial effects:

[0031] 1. The present invention can greatly improve the efficiency and accuracy of urban governance through multidimensional data classification, analysis and visualization methods; first, the multidimensional classification of data (such as geographical areas, functional areas, time dimensions, etc.) can help decision makers understand the complexity of the city from multiple perspectives and reveal potential correlations and patterns; secondly, correlation analysis (such as Pearson correlation coefficient, Spearman rank correlation coefficient and chi-square test) makes the relationship between different types of data clearer, helping to identify key influencing factors and trends; finally, through the visualization module, the classified data and analysis results can be presented intuitively, assisting decision makers to quickly grasp information, identify risk points and optimize resource allocation, and promote more refined and intelligent urban management and services; this method not only enhances the utilization value of data, but also effectively improves the accuracy and response speed of urban governance, and promotes the construction of smart cities.

[0032] 2. The present invention greatly improves the data insight ability and decision-making efficiency in urban governance through a multi-dimensional visualization analysis method; by calculating the strength, centrality, heterogeneity and relationship anomaly coefficient of the nodes, the risk level of each node can be comprehensively evaluated, and refined management can be performed based on the risk level; the system uses a dynamic and highly interactive network diagram to intuitively display the correlation and risk status between nodes, helping managers to quickly identify high-risk areas and take corresponding intervention measures; hierarchical views and real-time update functions further improve the operability of the data, supporting decision makers to conduct multi-angle analysis from macro to micro, from static to dynamic; in addition, combined with the abnormal behavior analysis of the nodes, potential risks can be accurately predicted and resource allocation can be optimized, thereby improving the response speed and accuracy of urban management and promoting the construction of smart cities and sustainable development. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] In order to facilitate understanding by those skilled in the art, the present invention is further described below with reference to the accompanying drawings.

[0034] Figure 1 It is a principle block diagram of the present invention. DETAILED DESCRIPTION

[0035] The technical solution of the present invention will be clearly and completely described below in conjunction with the embodiments. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0036] It should be understood that the terms “include” and “comprising” used in the specification and claims of the present disclosure indicate the presence of described features, integers, steps, operations, elements and / or components, but do not exclude the presence or addition of one or more other features, integers, steps, operations, elements, components and / or collections thereof.

[0037] It should also be understood that the terms used in this disclosure are only for the purpose of describing specific embodiments and are not intended to limit the disclosure. As used in this disclosure and claims, the singular forms of "a", "an", and "the" are intended to include the plural forms unless the context clearly indicates otherwise. It should also be further understood that the term "and / or" used in this disclosure and claims refers to any combination of one or more of the associated listed items and all possible combinations, including these combinations.

[0038] See also Figure 1 As shown, a GIS-based smart city refined management platform includes: a GIS acquisition module, a multidimensional analysis and conversion module, a multidimensional visualization module, a visualization analysis module and a visualization report module;

[0039] The GIS acquisition module is used to collect multi-source data of the city, including transportation, environment, energy, and public safety, and integrate the multi-source data, combining the spatial data in the geographic information system (GIS) with non-spatial data such as sensors and social media to obtain the city data set; and the city data set is processed by deduplication, denoising, and standardization to ensure the accuracy and consistency of the data;

[0040] The multidimensional analysis transformation module is used to classify urban data sets, including: geographic area classification (based on the natural geographical features of the city, such as topography, climate, water system and land use type), management area classification (based on the administrative division and management structure of the city, such as administrative divisions, functional management areas, communities and service areas, etc.), functional area classification (classified according to the functional characteristics of each area of ​​the city, such as commercial areas, residential areas, industrial areas and educational areas, etc.), time dimension classification (classifying data based on time, such as dividing data by hours, days, months, seasons, etc.), socio-economic classification (classified according to the economic conditions and social development levels of urban areas, etc.), and socio-economic classification (classified according to the economic conditions and social development levels of urban areas, etc.). Classification (such as high-income areas, low-income areas and developing areas), environmental quality classification (based on environmental data, such as air quality, noise pollution, water quality, etc.), infrastructure classification (based on the distribution and construction of urban infrastructure, such as transportation facilities, water supply systems and power systems, etc.), safety risk classification (based on public safety data, which may include high-risk fire areas, high-incidence traffic accident areas and high-crime areas, etc.) and service coverage classification (based on the distribution of social service facilities, such as medical service coverage areas, educational resource coverage areas and community services, etc.); the correlation analysis of several types of classified data is carried out, specifically:

[0041] Obtain the corresponding data in the two data types after classification. The two corresponding data are data A and data B for calculation: It should be noted that each data contains several variables, and each variable may represent the data of a time point or a region;

[0042] The linear relationship between data A and data B is analyzed by Pearson correlation coefficient: According to the formula: Output the Pearson correlation coefficient r between x and y; where x and y are the variable values ​​in data A and data B respectively (data in two regions or time points); n is the number of data points; ∑xy is the sum of the products of each pair of corresponding data in data A and data B; ∑x and ∑y are the sums of all data points in data A and data B respectively; the value range of the Pearson correlation coefficient r is between -1 and 1: when r=1, it indicates a complete positive correlation, when r=−1, it indicates a complete negative correlation, and r=0, it indicates no linear correlation;

[0043] If the data is ranking data (the ranking of a certain area and the ranking of another area), use the Spearman rank correlation coefficient to calculate: by the formula: Output the Spearman rank correlation coefficient ρ between x and y, d i is the ranking difference of the i-th pair of data, , n is the number of data points; for each variable (such as x and y), the data points are ranked in order of size; the smallest value is ranked 1, the second smallest value is ranked 2, and so on; if there are repeated values, they are assigned the same rank and calculated as the average; for example, if x=[3,1,4,2], then the rank is [2,1,4,3]; the value range of the Spearman coefficient is the same as that of the Pearson, -1 to 1, and the closer the value is to 1 or -1, the stronger the monotonic relationship;

[0044] If there are two classified data sets (traffic safety data and crime rate data divided by region), use the chi-square test to calculate: by formula: Output the chi-square statistic between x and y , where O i is the observed frequency, that is, the frequency of the data actually observed; E i is the expected frequency, that is, the expected value calculated based on the marginal probability under the assumption that the two variables are independent, and E i = (row sum) × (column sum) / total number of samples; it should be noted that: Chi-square statistic Compare with the critical value to determine whether there is a significant correlation between two categorical data sets; the critical value of chi-square Depends on the degrees of freedom and significance level; the formula for calculating the degrees of freedom is: degrees of freedom = (r1-1) × (c1-1), r1 is the number of rows (the number of categories of categorical variable 1), c1 is the number of columns (the number of categories of categorical variable 2); then look up the chi-square distribution table based on the degrees of freedom and significance level to get the critical value ; When the significance level is 0.05, the critical value can be found from the chi-square distribution table; then the calculated chi-square statistic And the critical value found For comparison: If the chi-square statistic >Threshold , then reject the null hypothesis, there is a significant correlation between the two categorical variables; if the chi-square statistic <Threshold value , then the null hypothesis cannot be rejected and there is no significant correlation between the two categorical variables;

[0045] The multidimensional visualization module is used to obtain various classified data and display them visually:

[0046] Obtain the classified data set and its correlation analysis results. First, filter the data. After data cleaning and filtering, map the classified data into a network diagram for intuitive display. The classified data is presented in the form of nodes. The color and size of the nodes can reflect the characteristics such as the number or weight of the data. The positively correlated data nodes are connected by solid lines, and the color and thickness of the line indicate the strength of the correlation. The negatively correlated nodes are connected by dotted lines, and visual optimization is also performed based on the strength of the correlation. The data nodes without correlation remain independent and are not connected.

[0047] The visualization analysis module is used to analyze the data after visualization connection, specifically:

[0048] By calculating the strength of each node, that is, the sum of all edge weights of the node, through the formula: Output node edge weight sum ,in, Indicates the strength of the correlation between nodes i1 and j1 (positive or negative correlation); then calculate the centrality of the node in the network, select different centrality metrics, and the centrality coefficient: indicates the number of edges directly connected to the node. ; Betweenness centrality coefficient: indicates the importance of a node in the shortest path, the formula is: Output betweenness centrality coefficient ,in, is the number of shortest paths from node s1 to t1, is the number of paths passing through node i1; then analyze whether the relationship between the node and other nodes is balanced, through the formula: Get the node heterogeneity coefficient ,when The closer it is to 0, the more positive and negative correlations of the node are balanced, and vice versa, it means that the correlation of the node tends to a certain relationship (positive or negative);

[0049] Next, we analyze the importance and heterogeneity of nodes:

[0050] First, calculate the deviation between the node strength and the mean strength of all nodes, using the formula: Output deviation ,in, is the mean value of all node strengths, is the standard deviation; when When a certain threshold is exceeded, it indicates that the strength of the node is abnormal;

[0051] Then calculate whether the correlation between the node and other nodes is abnormal, using the formula: Output relation anomaly coefficient ,in, is the mean of the edge weights between all nodes; the higher Indicates that the relationship network of the node is unbalanced and there is potential risk;

[0052] The calculated relationship anomaly coefficient , degree of deviation and betweenness centrality coefficient Perform comprehensive calculations using the set formula: Output risk abnormality coefficient , where b1, b2 and b3 are preset weight parameters;

[0053] The visual reporting module is used to obtain the risk anomaly coefficient Carry out refined management, specifically:

[0054] Risk Abnormality Coefficient Perform data hierarchical management: Compared with the preset threshold, the preset threshold includes threshold T1 and threshold T2, and threshold T1 < threshold T2; if the risk abnormality coefficient < threshold T1, it is a low-risk layer, the risk of the node is low and can be ignored or monitored; if threshold T1 < risk anomaly coefficient < threshold T2, it is a medium risk layer, indicating that the node has potential risks and needs to be paid attention to and preventive measures should be taken; if the risk abnormality coefficient ≥Threshold T2, it is a high-risk layer, indicating that the node is at high risk and requires immediate intervention or adjustment.

[0055] The multi-dimensional visualization module not only obtains and displays the status of different risk levels (low risk layer, medium risk layer, high risk layer), but also highlights high-risk nodes by combining visual elements such as color, size, and shape to help managers quickly identify key areas that need attention. For example, when the traffic accident risk factor of a certain urban area exceeds the threshold T2, the system will highlight the traffic nodes in the area in red and enlarge their size to remind managers that immediate intervention measures are needed in the area. At the same time, low-risk areas will be marked in green and kept small in size, indicating that they are low-risk and can be ignored or monitored routinely.

[0056] The multi-dimensional visualization module also supports dynamic display functions, which can update the network diagram in real time according to the time dimension and show the risk changes between nodes. For example, if the correlation trend between air quality and traffic flow in a certain area is displayed, the system can display the changes in traffic flow in real time when the air quality deteriorates, helping managers to gain insight into the interactive impact of the environment and traffic. When the risk score or status of a node changes, the graph will automatically update to reflect the new network structure and risk distribution in a timely manner. In addition, the multi-dimensional visualization module has powerful interactive functions, allowing users to freely zoom, drag, filter and highlight nodes and connections of interest. For example, managers can choose to view only the education resources within a specific time period. The negative correlation between the source and the crime rate. The correlation value, node category or specific time trend can be viewed through the hover function to deeply understand the correlation behind these data. The multi-dimensional visualization module also supports hierarchical views, which display information at different levels step by step. For example, the first-level view displays the network structure of the entire city, and users can get a macro overview; the second-level view focuses on high-risk areas, showing detailed information on the nodes in these areas and their risk levels, helping managers to understand the risk distribution more carefully; the third-level view displays in-depth information on specific nodes, such as key indicators such as node strength, centrality, and relationship anomaly coefficient, providing more in-depth analysis tools to help make accurate decisions.

[0057] The preferred embodiments of the present invention disclosed above are only used to help explain the present invention. The preferred embodiments do not describe all the details in detail, nor do they limit the invention to only specific implementation methods. Obviously, many modifications and changes can be made according to the content of this specification. This specification selects and specifically describes these embodiments in order to better explain the principles and practical applications of the present invention, so that those skilled in the art can understand and use the present invention well. The present invention is limited only by the claims and their full scope and equivalents.

Claims

1. A GIS-based smart city refined management platform, including a multidimensional analysis conversion module, a multidimensional visualization module, a visualization analysis module and a visualization report module; characterized in that: The multidimensional analysis conversion module is used to classify the urban data set to obtain the classified data, and then perform correlation analysis to obtain the Pearson correlation coefficient r, Spearman rank correlation coefficient ρ and square statistic between the two data. ; The visualization module is used to obtain various classified data for visualization; The visualization analysis module is used to analyze the data after visualization connection, specifically: Calculate the strength of each node to get the sum of the node edge weights ; Calculate the centrality of the node in the network to get the centrality coefficient and betweenness centrality coefficient ; By analyzing the number of paths to node i1, we can determine whether the relationship between the node and other nodes is balanced and obtain the node heterogeneity coefficient. ; Then analyze according to the importance and heterogeneity of the nodes: calculate the deviation between the node strength and the mean strength of all nodes ; Then calculate whether the correlation between the node and other nodes is abnormal and output the abnormal coefficient of the relationship ; The calculated relationship anomaly coefficient , degree of deviation and betweenness centrality coefficient Comprehensive calculation to obtain the risk abnormality coefficient ; The visual report module is used to obtain the risk anomaly coefficient Carry out refined management, specifically: Risk Abnormality Coefficient Perform data hierarchical management: Compared with the preset threshold, if the risk abnormal coefficient < threshold T1, it is a low-risk layer; If the threshold T1 is less than the risk abnormality coefficient < threshold T2, it is a medium-risk layer; if the risk abnormality coefficient is ≥ threshold T2, it is a high-risk layer.

2. According to the GIS-based smart city refined management platform of claim 1, it is characterized in that: It also includes a GIS acquisition module; The GIS acquisition module is used to collect multi-source data of the city, including transportation, environment, energy and public safety, and integrate the multi-source data, combining the spatial data and non-spatial data in the geographic information system to obtain the city data set; and deduplicate, denoise and standardize the city data set to ensure the accuracy and consistency of the data.

3. The GIS-based smart city refined management platform according to claim 1, characterized in that: The specific process of performing correlation analysis to obtain the Pearson correlation coefficient r between two sets of data is: The two corresponding data are calculated as data A and data B, and each data contains several variables. Each variable may represent the data of a time point or a region. The linear relationship between data A and data B is analyzed by the Pearson correlation coefficient: From the formula: Output the Pearson correlation coefficient r between x and y; where x and y are the variable values ​​in data A and data B respectively; n is the number of data points; ∑xy is the sum of the products of each pair of corresponding data in data A and data B; and are the sum of all data points of data A and data B respectively.

4. According to claim 3, a GIS-based smart city refined management platform is characterized in that: The specific process of performing correlation analysis to obtain the Spearman rank correlation coefficient ρ between two sets of data is: If the data is ranked data, use the Spearman rank correlation coefficient to calculate: According to the formula: Output the Spearman rank correlation coefficient ρ between x and y, where di is the ranking difference of the i-th pair of data. , n is the number of data points.

5. According to claim 4, a GIS-based smart city refined management platform is characterized in that: Perform correlation analysis to obtain the chi-square statistic between the two sets of data The specific process is: Two categorical data sets, calculated using the chi-square test: Output the chi-square statistic between x and y by the formula: , where Oi is the observed frequency; Ei is the expected frequency, and Ei = (row sum) × (column sum) / total number of samples.

6. The GIS-based smart city refined management platform according to claim 1, characterized in that: The specific process of the visualization module for obtaining various classified data for visualization is as follows: Obtain the classified data set and its correlation analysis results, and map the classified data into a network diagram for intuitive display; the classified data is presented in the form of nodes, and the color and size of the nodes reflect the characteristics of the data quantity or weight, while the positively correlated data nodes are connected by solid lines, and the color and thickness of the line indicate the strength of the correlation; the negatively correlated nodes are connected by dotted lines, and visual optimization is also performed based on the strength of the correlation; the data nodes without correlation remain independent and are not connected.

7. The GIS-based smart city refined management platform according to claim 6 is characterized in that: The strength of each node is calculated to obtain the sum of the node edge weights The specific process is: Calculate the strength of each node, the sum of all edge weights of the node, by the formula: Output node edge weight sum , Represents the strength of the correlation between nodes i1 and j1.

8. The GIS-based smart city refined management platform according to claim 7 is characterized in that: The centrality of the computing node in the network is obtained by the centrality coefficient and betweenness centrality coefficient The specific process is: Calculate the centrality of a node in the network and select different centrality measures. The centrality coefficient is: the number of edges directly connected to the node is ; Betweenness centrality coefficient: indicates the importance of a node in the shortest path, the formula is: Output betweenness centrality coefficient , is the number of shortest paths from node s1 to t1, is the number of paths passing through node i1.

9. The GIS-based smart city refined management platform according to claim 8, characterized in that: The risk anomaly coefficient is obtained by comprehensive calculation The specific process is: Calculate the deviation between the node strength and the mean strength of all nodes using the formula: Output deviation ,in, is the mean value of all node strengths, is the standard deviation; then calculate whether the correlation between the node and other nodes is abnormal, through the formula: Output relation anomaly coefficient , is the mean of the edge weights between all nodes; the calculated relationship anomaly coefficient , degree of deviation and betweenness centrality coefficient Perform comprehensive calculations by setting the formula: Output risk abnormality coefficient , where b1, b2 and b3 are preset weight parameters.

Images