Network space potential metaphor map expression method

Through metaphorical map technology, based on network space link data, a mapping model between the position potential and administrative division map of the cyberspace is constructed, which solves the problem that existing technology is difficult to intuitively express the strength of the cyberspace and the relationship between the national neighbors, and achieves a more intuitive and accurate Cyberspace Position Map Expression.

CN118982694APending Publication Date: 2024-11-19Chinese People's Liberation Army Cyberspace Force Information Engineering University
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202410975219.7
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-07-19
Publication Date
2024-11-19

AI Technical Summary

Technical Problem

The existing technology is difficult to intuitively express the strength of cyberspace, accurately describe the national neighbor relationships in virtual spaces, and it is difficult to fully reflect the characteristics of "connection" of network logical relationships by using hierarchical relationships, and ignore countries such as network infrastructure, economic strength, and military strength. Hard power.

Method used

Using basic theories and methods of metaphor maps, based on basic data such as network space links, a metaphorical mapping model between the position potential of the network space and administrative division maps is constructed, and a network space is generated through the processes of graph embedding, graph clustering, spatial layout, map coloring, etc. Position metaphor map, map network link maps into two-dimensional planes using graph embedding algorithm, and optimizes national plate layout and coloring with K-means algorithm and force-guided layout algorithm.

Benefits of technology

It realizes a more intuitive and accurate expression of cyberspace strength, fully reflects the adjacent relationship between countries and network logical connections, and combines the comprehensive network strength index to ensure the visual effect of the map.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118982694A_ABST
    Figure CN118982694A_ABST
Patent Text Reader

Abstract

The invention discloses a network space potential metaphor map expression method, which comprises the following steps: constructing a node-link association graph based on network basic resource data, and generating a network link graph; the network basic resource data comprises an IP, an AS autonomous domain, a DNS server and vulnerabilities; mapping the network link graph into nodes on a two-dimensional plane by using a graph embedding algorithm; clustering discrete two-dimensional network nodes according to the country by using a K-means algorithm, and optimizing a clustering result in combination with modularity; calculating the network comprehensive strength index of each country, optimizing the layout of each country plate in the cyberspace through a force-guided layout algorithm based on the clustering result according to the AS direct connection condition of each country and the obtained network comprehensive strength index of each country, and carrying out coloring rendering on each country plate in the cyberspace. And finally forming a network space potential metaphor map. According to the method, the network space strength can be expressed more intuitively and accurately.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of network space visualization, and in particular to a network space potential metaphor map expression method. Background Art

[0002] In today's era of information explosion, cyberspace has become an indispensable part of human society. Cyberspace maps have become an important tool for understanding and recognizing cyberspace since the early days of cyberspace development. Through modeling, analysis and expression of cyberspace, using cyberspace maps, we can more clearly understand the complex relationship between network structure, data flow and information dissemination in cyberspace. The research on cyberspace maps has received attention and attention from the surveying and mapping and information fields, and has achieved a series of research results. Due to the virtualization and borderlessness of cyberspace, it is difficult to make cyberspace maps through traditional map projection, symbol design, map visualization and other methods. Metaphors are usually used to transform abstract cyberspace elements and phenomena into more intuitive and easier to understand forms, which can better assist people's understanding of cyberspace.

[0003] There have been corresponding studies on using metaphorical maps to describe cyberspace elements and information flows. For example, Chen et al. proposed the D-Map visualization method through metaphorical maps to simulate the process of information dissemination in social network space; WISEJ A used contour lines to metaphorically represent news maps, and used changes in lines to represent the importance or popularity of news events; Xin Rui et al. studied the spatial metaphorical expression and analysis of non-spatial data based on the Gosper curve and metaphorical map ideas; Liu Longhui et al. proposed a method for constructing a cyberspace metaphorical Gosper map and topographic map; Qi Kai et al. regarded resource nodes in cyberspace as the ontology, and used peaks and contour lines in traditional geographic space as the metaphor for visualization.

[0004] Cyberspace potential is mainly used to describe the potential influence and role of a country or region in network technology, information control, data flow, network security, etc. in the global cyberspace relative to other countries or regions. It reflects a country's competitiveness, influence in the global cyberspace, and its dominance and voice in international cyber affairs. Cyberspace potential is mainly used to characterize the network strength and network influence of different countries or regions. The current characterization and description of cyberspace potential is usually presented in the form of indicator model construction and thematic map visualization. The main methods for evaluating the comprehensive network strength of a country include the Belfer National Cyber ​​Power Index (NCPI), the Global Cybersecurity Index (GCI), the Cyber ​​Readiness Index 2.0 (CRI 2.0) and the "Cyber ​​Power Index" (CPI). By comprehensively considering the soft and hard power related to cyberspace, the network strength of each country is comprehensively evaluated.

[0005] By expressing the sphere of influence in cyberspace, we can reveal and analyze the influence, control ability, expansion trend and cooperation relationship of different countries in cyberspace, as well as their impact on the security, stability and development of cyberspace. For example, Leetaru et al. used Twitter data and conducted a visual analysis of Twitter activities around the world based on the Geographic Information System (GIS), revealing the usage patterns and activity of Twitter in different regions, thus reflecting the influence and sphere of influence of different regions in cyberspace.

[0006] In summary, the use of metaphors to model and express cyberspace environments has become an important way to visualize cyberspace maps. The description of cyberspace potential is a depiction and expression of the basic environment of cyberspace, and other thematic information related to cyberspace can be superimposed on the basic "base map". However, the current research on modeling and expressing cyberspace potential has the following shortcomings: (1) It is difficult to intuitively express cyberspace power and accurately describe the national neighboring relationship in virtual space by displaying the scope of cyberspace power through indicators, charts, etc.; (2) It is difficult to fully reflect the characteristics of "connection" of network logical relationships by using hierarchical relationship methods; (3) Network power is only expressed by the number and density of network connections, ignoring national hard power such as network infrastructure, economic strength, and military strength. Summary of the invention

[0007] In view of the above problems, the present invention proposes a method for expressing cyberspace potential metaphor map. With the help of basic theories and methods of metaphor maps, according to the basic characteristics of cyberspace, based on basic data such as cyberspace links, a metaphor mapping model between cyberspace potential and administrative division maps is constructed, and the cyberspace potential metaphor map is generated by using processes such as graph embedding, graph clustering, spatial layout, and map coloring. This map is used as the "base map" to realize the superimposed display of various cyberspace thematic elements.

[0008] In order to achieve the above object, the present invention adopts the following technical solutions:

[0009] A network space potential metaphor map expression method, comprising:

[0010] Step 1: construct a "node-link" association graph based on network basic resource data to generate a network link graph; the network basic resource data includes IP, AS autonomous domain, DNS server and vulnerability;

[0011] Step 2: Use graph embedding algorithm to map the network link graph into nodes on a two-dimensional plane;

[0012] Step 3: Cluster the discrete two-dimensional network nodes by country using the K-means algorithm, and optimize the clustering results based on modularity;

[0013] Step 4: Calculate the comprehensive network strength index of each country. Based on the clustering results of step 3, according to the direct connection of each country's AS and the obtained comprehensive network strength index of each country, optimize the layout of each country's cyberspace plate through the force-guided layout algorithm, and color and render each country's cyberspace plate, and finally form a potential metaphor map of cyberspace.

[0014] Furthermore, in step 2, the network link graph is mapped into nodes on a two-dimensional plane using the DeepWalk algorithm.

[0015] Furthermore, in step 3, the discrete two-dimensional network nodes are clustered in the following manner:

[0016] Randomly select k data points belonging to different countries as initial cluster centers; including: first randomly select an AS number as the first initial national cluster center, then for each AS number of the remaining countries that have not been selected, calculate the minimum distance between it and the existing cluster center, and select the next cluster center according to the probability distribution of the distance;

[0017] For each discrete two-dimensional network node, calculate its Euclidean distance to each cluster center and assign it to the cluster center closest to it;

[0018] For each cluster, recalculate its cluster center, and the new cluster center is the average value of all two-dimensional network nodes in the cluster;

[0019] The node allocation and cluster center update steps are iterated until the AS numbers of the same country belong to the same cluster and the AS numbers of different countries are not in the same cluster, and the clustering is terminated.

[0020] Furthermore, in step 3, optimizing the clustering results in combination with the modularity includes:

[0021] A bipartite graph is constructed based on the connection relationship after clustering. In the bipartite graph, each node corresponds to two vertices, which represent being in the current cluster and not being in the current cluster respectively.

[0022] For each node i, use the variable C i To store the cluster number, and then traverse each of the remaining nodes j, calculate the edge weight ω between i and j ij , and their cluster number C i and C j ;

[0023] If C i =C j , then ω ijAdd it to the modularity calculation, otherwise no modification is made, and finally the total modularity is calculated;

[0024] Finally, a pair of nodes with the largest modularity gain is selected from all node pairs, and they are merged into the same cluster. At the same time, the bipartite graph and modularity are updated to make the final clustering result optimal.

[0025] Furthermore, in step 4, the network comprehensive strength index of each country is calculated in the following manner:

[0026]

[0027] Among them, a j is the weight assignment of various indicators, M j Represents the indicators formed by network strength and overall strength respectively.

[0028] Furthermore, the indicators include the number of IPs, the number of AS autonomous domains, the number of DNS servers, the number of vulnerabilities, GDP and land area, among which the number of IPs, the number of AS autonomous domains, the number of DNS servers and the number of vulnerabilities belong to network strength indicators, and GDP and land area belong to overall strength indicators.

[0029] Furthermore, in step 4, based on the clustering results of step 3, according to the direct connection status of each country's AS and the obtained network comprehensive strength index of each country, the layout of each country's network space is optimized by the force-guided layout algorithm, including:

[0030] First, according to the clustering results, the country nodes of the same cluster are placed in the same area, and the position of each node is randomly initialized;

[0031] Secondly, the repulsive force, attractive force and connected force between nodes are calculated, and then the position of each node is updated according to the calculated force; then the force calculation and position update are repeated until the convergence condition is reached;

[0032] Finally, connection processing is carried out. According to the connection relationship between countries, connection lines or curves are drawn to represent the connection between countries. The area of ​​each country is allocated according to the calculated network comprehensive strength index, and the map is redrawn according to the updated coordinates.

[0033] Furthermore, in step 4, in order to ensure that the size of each country does not change when the graphic layout changes, the following operations are performed:

[0034] Define an objective function that measures the degree of polygon clustering and the size of empty spaces, which are non-adjacent parts of the map between countries:

[0035] F=ω1f1+ω2f2,

[0036]

[0037] S ij =|x i -x j ||y i -y j |,

[0038] Among them, F is the overall objective function, f1 is the objective function for measuring the degree of polygon aggregation, f2 is the objective function for measuring the size of the open space, ω1 and ω2 are the weight coefficients corresponding to f1 and f2 respectively, N is the number of polygons, ω ij is the weight coefficient between the i-th and j-th polygons, d ij is the distance between the i-th and j-th polygons, (x i ,y i )、(x j ,y j ) are the coordinates of the center points of the i-th and j-th polygons, S ij is the area between the circumscribed rectangles of the i-th and j-th polygons, A i and A j are the areas of the i-th and j-th polygons respectively;

[0039] Define the following constraints: The minimum distance between polygons is not less than a preset value d min ; The maximum area of ​​the open space does not exceed a preset value S max ; The area of ​​each polygon after updating the coordinates is equal to the original area;

[0040] The objective function and constraints are combined into an optimization problem, and nonlinear programming is used to solve the optimization problem and obtain the optimal solution; finally, the coordinates of the relevant points and lines are updated through the calculated optimal solution point.

[0041] Compared with the prior art, the present invention has the following beneficial effects:

[0042] The present invention proposes a metaphorical map expression method for cyberspace potential. With the help of basic theories and methods of metaphorical maps, according to the basic characteristics of cyberspace, based on basic data such as cyberspace links, a metaphorical mapping model between cyberspace potential and administrative division maps is constructed. The cyberspace potential metaphorical map is generated by processes such as graph embedding, graph clustering, spatial layout, and map coloring. This map is used as a "base map" to realize the superposition display of various cyberspace thematic elements. The present invention converts "virtual, borderless" cyberspace links into "physical, bounded" cyberspace potential maps, which can express the strength of cyberspace more intuitively and accurately. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] Figure 1An example diagram of a metaphor of similarity between cyberspace potential and map content provided by an embodiment of the present invention;

[0044] Figure 2 A flowchart of a network space potential metaphor map expression method provided by an embodiment of the present invention;

[0045] Figure 3 A flowchart of graph embedding using the DeepWalk algorithm provided in an embodiment of the present invention;

[0046] Figure 4 A flow chart of the steps of the K-means algorithm provided in an embodiment of the present invention;

[0047] Figure 5 A schematic diagram of an open space area provided by an embodiment of the present invention;

[0048] Figure 6 A schematic diagram of force-directed layout algorithm optimization provided by an embodiment of the present invention;

[0049] Figure 7 A schematic diagram of a cyberspace potential metaphor map constructed for an embodiment of the present invention. DETAILED DESCRIPTION

[0050] The present invention will be further explained below with reference to the accompanying drawings and specific embodiments:

[0051] "Metaphor" is a concept in linguistics. The core of metaphor is to deepen people's knowledge and understanding of unfamiliar things by unconventionally juxtaposing familiar and unfamiliar things. Map metaphor is a method of visually expressing abstract data using maps as carriers. By transferring the meaning of the subject and the metaphor based on the similarity between the subject and the metaphor, the metaphor can well display the attributes of the subject at the visual level.

[0052] Cyber ​​power is a concept proposed with the help of "geo-position". Geo-position is used to measure the influence and strategic advantages of a country or region on the international stage, which are determined by its geographical location, resources, population, economy, military strength, political stability and other factors. With the understanding of geo-position, "cyber power" refers to the potential influence and ability of a country or region in network technology, information control, data flow, network security and other aspects relative to other countries or regions in the global cyberspace. It reflects a country's competitiveness, influence in the global cyberspace, and its dominance and voice in international cyber affairs. Cyber ​​power usually refers to the ability of a country, organization or individual to control information and data in cyberspace and to use cyberspace to influence, control or attack. Cyber ​​power includes both the ability in network defense and security, and the ability to use network technology and information resources to influence others at the economic, political and social levels. Cyber ​​power is usually soft power. How to intuitively express the abstract and difficult-to-describe cyber power in an easy-to-understand map form.

[0053] The borders, outlines and areas between countries in the map can clearly show the borders, locations and territorial areas of the countries. Generally, people can have an intuitive understanding of the geographical space through maps. For this reason, the present invention considers metaphorically representing the sphere of influence of the country in the cyberspace through the area, borders and locations of the country in the geographical space. Figure 1As shown in the figure, the ontology "cyberspace potential" mainly includes the basic understanding of the cyberspace environment, and it is necessary to understand the scope of influence of cyberspace in each country, the network connectivity and key nodes. The metaphor "administrative division map" mainly reflects the territorial scope, administrative division status, location of administrative center and traffic conditions of a country or region. Although the two belong to different spatial affiliations, their contents have strong similarities, which provides a basis for the metaphorical expression of the cyberspace potential map. The form of administrative division map can be used to express the sphere of influence of cyberspace. Selecting and highlighting the characteristics of the metaphor to creatively interpret the ontology is the key to the metaphor map truly constituting a visual metaphor (the basic theoretical problem of metaphor map). The metaphor "administrative division map" is composed of various types of settlements, roads, water systems, administrative districts and other elements. The administrative area represents the size of the geographical territory occupied by each administrative division, the boundary represents the geographical territory demarcated by the geographical space, and the road represents the interconnection network between various regions, reflecting the traffic accessibility between cities. Hub cities reflect their key position in the field of transportation. The ontology "cyberspace potential map" needs to express the network sphere of influence of each country, the influence boundary of network power, the network connectivity between countries, the key nodes of network connection, etc. Through the similarity fusion of map metaphors, the similarity is highlighted, and finally the materialization process of metaphor map visual expression needs to be realized through symbolization. The network sphere of influence is represented as a surface symbol with different colors, the network backbone network is represented as a road network linear symbol, and the AS (Autonomous System) level, POP (Point of Presence) level, routing level and IP level network backbone connection are represented as different road classifications. Visual variables such as color and thickness are fully used to depict different levels of network connection, and the key nodes of the network are represented as point symbols of different levels.

[0054] like Figure 2 As shown, the present invention proposes a method for expressing a cyberspace potential metaphor map. The cyberspace potential metaphor map is a method for making a cyberspace map that uses the cyberspace sphere of influence as the main body, the geographical map as the metaphor, and the geographical map elements that are familiar to people to express the complex and difficult-to-understand cyberspace sphere of influence. The basic idea is to convert the "virtual, borderless" cyberspace links into a "physical, bounded" cyberspace potential map. The main steps include:

[0055] (1) Network node-link data graphing. Based on network basic resource data, such as AS, DNS, IP, and vulnerability data, a "node-link" association graph is constructed to generate a network (space) link graph;

[0056] (2) Mapping the network link graph into a two-dimensional plane. Using the graph embedding algorithm, the high-dimensional network space "graph" is mapped into nodes on a two-dimensional plane, similar to the "map projection" of geographic space;

[0057] (3) Clustering of network nodes based on country. The discrete two-dimensional network nodes are clustered according to country using the K-means algorithm, and the clustering results are optimized in combination with modularity;

[0058] (4) National power plate layout based on force-guided layout. Based on the cyberspace potential index system, the network potential of each country is calculated, and then the network comprehensive strength index of each country is calculated. Based on the clustering results, according to the direct connection of each country's AS and the obtained network comprehensive strength index of each country, the layout of each national plate in cyberspace is optimized through the force-guided layout algorithm. After that, each national plate in cyberspace is colored and rendered, and finally a metaphorical map of cyberspace potential is formed.

[0059] The method specifically includes:

[0060] 1 Graph Embedding Algorithm

[0061] Graph Embedding algorithm is a process of mapping graph data (usually high-dimensional dense matrix) into low-dimensional dense vector. Network AS connection data is high-dimensional graph connection data. The present invention uses DeepWalk algorithm to achieve dimensionality reduction mapping from network link graph data to two-dimensional plane. DeepWalk is a graph structure data mining algorithm that combines Random Walk and Word2vec algorithms. The main steps include (such as Figure 3 shown):

[0062] (1) Random walk sampling node sequence;

[0063] (2) Use the Skip-gram model to learn the representation vector and perform random walk sampling from each node in the AS connected network to obtain locally correlated data;

[0064] (3) Skip-gram training is then performed on the sampled data to obtain the vectorized representation of the nodes, maximize the node co-occurrence, and use Hierarchical Softmax to make a classifier for ultra-large-scale classification;

[0065] (4) Compute the embedding of each node in the graph.

[0066] The following is the pseudo code for embedding AS connectivity data into a two-dimensional plane using the DeepWalk algorithm:

[0067]

[0068] 2 Network Space Graph Clustering Algorithm

[0069] 2.1 K-means algorithm

[0070] The K-means algorithm is an iterative clustering analysis algorithm. Its core idea is to divide n objects in a data set into k clusters so that the sum of the distances from each object to the center (or mean point, centroid) of its cluster is minimized, where the distance usually refers to the Euclidean distance. Figure 4 )include:

[0071] (1) Randomly select k data points belonging to different countries as the initial cluster centers and name them with "country names". First, randomly select an AS number as the first initial national cluster center. Then, for each unselected AS number (of other countries), calculate the minimum distance between it and the existing cluster center, and select the next cluster center according to the probability distribution of the distance. In this way, the K-means algorithm can make the distance between the initial cluster centers farther, thus avoiding falling into the local optimal solution.

[0072] (2) Calculate the distance between each node and the cluster center. For each data point in the data set, calculate its Euclidean distance to each cluster center and assign it to the cluster center closest to it. The calculation formula is as follows:

[0073]

[0074] Where x is the AS point, c i is the i-th cluster center, that is, the i-th country, d is the data dimension (usually two-dimensional), x j and c ij are x and c respectively i The value in the j dimension.

[0075] (3) Recalculate the cluster center. For each cluster (country), recalculate its cluster center. The new cluster center is the average value of all data points in the cluster. The calculation formula is as follows:

[0076]

[0077] Where S i is the set of data points of the ith cluster (country), |S i | is the number of data points in the set (number of AS numbers).

[0078] (4) Clustering iteration: Iterate the allocation and update steps until the AS numbers of the same country belong to the same cluster (country), and the AS numbers of different countries are not in the same cluster.

[0079] 2.2 Modularity Optimization Algorithm

[0080] Modularity is used to measure the quality of the network structure division results of the evaluation algorithm. Therefore, the modularity of the above clustering results can be calculated, and then the compactness within the clustering and the stability of the clustering results can be enhanced by optimizing the modularity of the clustering results. The calculation formula of modularity is:

[0081]

[0082] Among them, Q is modularity. The larger the modularity, the higher the quality of the clustering result. The range of Q is [-0.5, 1]. When the value of Q is between 0.3-0.7, it means that the clustering effect is very good. m is the number of edges in the graph, A ij represents the element in the adjacency matrix A, ω represents the edge weight between node i and node j, k i and k j Respectively represent the number of nodes adjacent to node i and node j. When i and j belong to the same country, &(c i , c j )=1, when i and j do not belong to the same country, &(c i , c j )=0.

[0083] First, a bipartite graph is constructed based on the connection relationship after clustering. In the bipartite graph, each node corresponds to two vertices, which represent the nodes in the current cluster and those not in the current cluster respectively. Second, for each node i, the variable C is used i To store the cluster number it is in, and then traverse each of the remaining nodes j to calculate the edge weight ω between i and j ij , and their cluster number C i and C j ; if C i =C j , then ω ij Add it to the modularity calculation, otherwise no modification is made, and finally the total modularity is calculated by formula Q. Finally, a pair of nodes with the largest modularity gain is selected from all node pairs, and they are merged into the same cluster. At the same time, the bipartite graph and modularity are updated to make the final clustering result optimal.

[0084] 3 Network space layout algorithm

[0085] 3.1 Network space potential indicators

[0086] The number of network ASs, links, IPs, and vulnerabilities can reflect the completeness of cyberspace facilities and the communication capabilities of network links to a certain extent, but cannot fully reflect the cyberspace potential. In order to fully reflect the comprehensive strength, influence, control ability, expansion trend, and cooperative relationship with other countries of different countries in cyberspace, the cyberspace potential should not only consider the hard power such as network technology strength and network security ability, but also the soft power such as information control and dissemination ability, international cooperation and diplomacy, and cyberspace laws, regulations, and policy environment.

[0087] The present invention constructs a network space potential index system model based on the NCPI index, comprising the following steps:

[0088] (1) Factors influencing cyberspace potential: including cyber technology strength, cyber security capabilities, international cooperation and diplomacy, economic strength, military strength, population size, and scientific and technological level;

[0089] (2) Design indicators: For each key factor, design corresponding indicators to quantify its influence. For example, economic strength is measured by indicators such as gross domestic product (GDP), trade volume, and foreign exchange reserves; military strength is measured by the number of armed forces personnel and the ratio of military expenditure to GDP; scientific and technological level is measured by scientific and technological journal articles, the number of patent applications, education penetration rate, and the export volume of high-tech industries. The specific measurement factors are shown in Table 1.

[0090] Table 1 Quantitative indicators of key factors

[0091]

[0092] Drawing on the geo-potential model, a country’s cyber strength can also be divided into the cyberspace influence part (soft power) and the country’s overall influence part (hard power). Therefore, the weights assigned to the comprehensive cyber strength index are shown in Table 2.

[0093] Table 2 Network comprehensive strength index

[0094]

[0095]

[0096] (3) After assigning weights to the data, the Z-score standardization method is used to standardize the data to between (0, 1]. The specific steps are as follows:

[0097] Transform the sequence x1, x2, ..., n:

[0098] in,

[0099] Then the new sequence y1, y2, ...yn The mean is 0 and the variance is 1. Finally, the comprehensive strength of each country is calculated by the following formula.

[0100]

[0101] In the formula, a j is the weight assignment of various indicators, M j Represents the indicators formed by network strength and overall strength respectively.

[0102] 3.2 Force-directed layout algorithm

[0103] The clustering of network connections based on national entities fully considers the number of ASs in the networks between countries and the direct associations between ASs, but it is difficult to accurately describe the "neighboring relationship" in cyberspace. For example, the more network connections there are between countries, the higher the degree of dependence of the network communications between the two countries, which can be expressed by "neighboring relationships". Therefore, the present invention adopts a force-guided layout algorithm to optimize the layout of the "plates" of national cyberspace. The force-directed layout algorithm (Force-Directed Layout) is based on the mechanical principles in physics and simulates the repulsion and attraction between nodes to achieve the optimal node position layout. The basic process is:

[0104] First, according to the clustering results, the country nodes of the same cluster are placed in the same area, and the position of each node is randomly initialized;

[0105] Secondly, calculate the repulsive force, attractive force and connected force between nodes. The repulsive force makes the national nodes in the same cluster keep a certain distance from each other to avoid overlap; the attractive force attracts the national nodes in the same cluster to shorten the distance to enhance the clustering effect; the connecting force adds connecting lines between national nodes in different clusters and applies a certain attractive force on these connecting lines to ensure that the connection between countries is not too weak; then update the position of each node (country) according to the calculated force; then repeat the force calculation and position update until the convergence condition is reached;

[0106] Finally, the connection processing is carried out. According to the connection relationship between countries, connection lines or curves are drawn to represent the connection between them, and the comprehensive network strength of each country is added. The area of ​​each country is allocated according to the calculated comprehensive strength index (specifically, the area size is allocated according to the proportion of the comprehensive network strength index of all countries), and the map is redrawn according to the updated coordinates. Among them, in order to ensure that the area size of each country does not change when the graphic layout changes, and is still determined by its comprehensive network strength, the following operations need to be performed:

[0107] Define the objective function: First, define an objective function that measures the degree of polygon aggregation and the size of empty space. Empty space is the non-adjacent part between countries in the map, such as Figure 5 shown.

[0108] Consider using the distance between polygons to represent the degree of aggregation, and calculating the area of ​​vacant land to represent the size of vacant land. Let the objective function of polygon aggregation be f1, the objective function of vacant land size be f2, and the overall objective function be F = ω1f1 + ω2f2, where ω1 and ω2 are weight coefficients used to balance the importance of the two objective functions; secondly, define the distance between polygons as the Euclidean distance between the center points of each polygon. Assume that the coordinates of the center point of the i-th polygon are (x i ,y i ), the distance between polygons can be expressed as:

[0109]

[0110] It is assumed that the polygon has n vertices, whose coordinates are (x1, y1), (x2, y2)... (x n ,y n ), then the coordinates of the polygon center point are (x, y): The objective function of polygon clustering can be defined as the weighted sum of the distances between polygons. Where N is the number of polygons, ω ij is the weight coefficient between the i-th and j-th polygons. The objective function of the open space size can be defined as the sum of the areas between the polygon circumscribed rectangles minus the sum of the areas of all polygons:

[0111]

[0112] Among them, S ij is the area between the circumscribed rectangles of the i-th and j-th polygons, A i and A j are the areas of the i-th and j-th polygons respectively, and the polygon area is:

[0113]

[0114] Where n is the number of polygon vertices, (x i ,y i ) is the coordinate of the i-th vertex, (x i+1 ,y i+1 ) are the coordinates of the next vertex. The area between the external matrices is:

[0115] S ij =|x i -x j ||y i-y j |

[0116] Among them, (x i ,y i ) and (x j ,y j ) are the center coordinates of the i-th and j-th polygons respectively.

[0117] (2) Define constraints: To ensure that the polygons are clustered together and there are no empty spaces, add the following constraints:

[0118] ①The minimum distance between polygons is not less than a preset value d min ;

[0119] ②The maximum area of ​​the open space does not exceed a preset value S max ;

[0120] ③The area of ​​each polygon after updating the coordinates needs to be equal to the original area.

[0121] (3) Solving the optimization problem: Combining the objective function and constraints into an optimization problem, we can get the following standard optimization problem:

[0122] minimize F(x)

[0123] subject togi(x)≤0,i=1,...,m

[0124] hj(x)=0,j=1,...,p

[0125] Among them, x is the coordinate vector of the polygon, gi(x) and hj(x) are the inequality constraint and equality constraint functions respectively, m and p are the number of inequality constraints and equality constraints respectively, nonlinear programming is used to solve the optimization problem and obtain the optimal solution; finally, the coordinates of the relevant points and lines are updated through the calculated optimal solution point. The overall process is shown as follows Figure 6 shown.

[0126] To verify the effect of the present invention, the following experiment was performed:

[0127] 1) Experiment on constructing network potential metaphor map

[0128] The data used in the experiment include: AS connectivity data, vulnerability data of various countries, number of IPs in various countries, number of ASNs, DNS servers, domain names and other data.

[0129] The number of IP addresses, number of AS autonomous domains, number of DNS servers, number of domain names, Internet penetration rate, number of secure Internet servers, number of vulnerabilities, GDP, and land area of ​​each country are used as indicators of national network influence. The specific units and deadlines are shown in Table 3.

[0130] Table 3 Network comprehensive strength index indicators

[0131]

[0132] The calculation formula in 3.1 can be used to obtain the comprehensive network strength index of each country, and the top ten countries are shown in Table 4.

[0133] Table 4 Top 10 countries in terms of comprehensive network strength

[0134]

[0135] 2) Visualization results of network potential metaphor map

[0136] Use GMap to directly visualize the AS connectivity data, change the clustering basis to "country name", and use the force-directed layout algorithm and add the network potential to generate a network potential metaphor map as shown below: Figure 7 shown.

[0137] The above is only a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principle of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.

Claims

1. A network space potential metaphor map expression method, characterized in that: include: Step 1: construct a "node-link" association graph based on network basic resource data to generate a network link graph; the network basic resource data includes IP, AS autonomous domain, DNS server and vulnerability; Step 2: Use graph embedding algorithm to map the network link graph into nodes on a two-dimensional plane; Step 3: Cluster the discrete two-dimensional network nodes by country using the K-means algorithm, and optimize the clustering results based on modularity; Step 4: Calculate the comprehensive network strength index of each country. Based on the clustering results of step 3, according to the direct connection of each country's AS and the obtained comprehensive network strength index of each country, optimize the layout of each country's cyberspace plate through the force-guided layout algorithm, and color and render each country's cyberspace plate, and finally form a potential metaphor map of cyberspace.

2. A method for expressing a cyberspace potential metaphor map according to claim 1, characterized in that: In step 2, the network link graph is mapped into nodes on a two-dimensional plane using the DeepWalk algorithm.

3. The method for expressing a cyberspace potential metaphor map according to claim 1, characterized in that: In step 3, the discrete two-dimensional network nodes are clustered in the following manner: Randomly select k data points belonging to different countries as initial cluster centers; including: first randomly select an AS number as the first initial national cluster center, then for each AS number of the remaining countries that have not been selected, calculate the minimum distance between it and the existing cluster center, and select the next cluster center according to the probability distribution of the distance; For each discrete two-dimensional network node, calculate its Euclidean distance to each cluster center and assign it to the cluster center closest to it; For each cluster, recalculate its cluster center, and the new cluster center is the average value of all two-dimensional network nodes in the cluster; The node allocation and cluster center update steps are iterated until the AS numbers of the same country belong to the same cluster and the AS numbers of different countries are not in the same cluster, and the clustering is terminated.

4. The method for expressing a cyberspace potential metaphor map according to claim 1, characterized in that: In step 3, optimizing the clustering results in combination with the modularity includes: A bipartite graph is constructed based on the connection relationship after clustering. In the bipartite graph, each node corresponds to two vertices, which represent being in the current cluster and not being in the current cluster respectively. For each node i, use the variable C i To store the cluster number, and then traverse each of the remaining nodes j, calculate the edge weight ω between i and j ij , and their cluster number C i and C j ; If C i =C j , then ω ij Add it to the modularity calculation, otherwise no modification is made, and finally the total modularity is calculated; Finally, a pair of nodes with the largest modularity gain is selected from all node pairs, and they are merged into the same cluster. At the same time, the bipartite graph and modularity are updated to make the final clustering result optimal.

5. The method for expressing a cyberspace potential metaphor map according to claim 1 is characterized in that: In step 4, the network comprehensive strength index of each country is calculated in the following manner: Among them, a j is the weight assignment of various indicators, M j Represents the indicators formed by network strength and overall strength respectively.

6. A method for expressing a cyberspace potential metaphor map according to claim 5, characterized in that: The indicators include the number of IPs, the number of AS autonomous domains, the number of DNS servers, the number of vulnerabilities, GDP and land area. Among them, the number of IPs, the number of AS autonomous domains, the number of DNS servers and the number of vulnerabilities are network strength indicators, and GDP and land area are overall strength indicators.

7. The method for expressing a cyberspace potential metaphor map according to claim 1, characterized in that: In step 4, based on the clustering results of step 3, according to the direct connection status of each country's AS and the obtained network comprehensive strength index of each country, the layout of each country's network space is optimized by the force-guided layout algorithm, including: First, according to the clustering results, the country nodes of the same cluster are placed in the same area, and the position of each node is randomly initialized; Secondly, the repulsive force, attractive force and connected force between nodes are calculated, and then the position of each node is updated according to the calculated force; then the force calculation and position update are repeated until the convergence condition is reached; Finally, connection processing is carried out. According to the connection relationship between countries, connection lines or curves are drawn to represent the connection between countries. The area of ​​each country is allocated according to the calculated network comprehensive strength index, and the map is redrawn according to the updated coordinates.

8. The method for expressing a cyberspace potential metaphor map according to claim 1, characterized in that: In step 4, in order to ensure that the size of each country does not change when the graphic layout changes, the following operations are performed: Define an objective function that measures the degree of polygon clustering and the size of empty spaces, which are non-adjacent parts of the map between countries: F=ω1f1+ω2f2, S ij =|x i -x j ||and i -and j |, Among them, F is the overall objective function, f1 is the objective function for measuring the degree of polygon aggregation, f2 is the objective function for measuring the size of the open space, ω1 and ω2 are the weight coefficients corresponding to f1 and f2 respectively, N is the number of polygons, ω ij is the weight coefficient between the i-th and j-th polygons, d ij is the distance between the i-th and j-th polygons, (x i ,y i )、(x j ,y j ) are the coordinates of the center points of the i-th and j-th polygons, S ij is the area between the circumscribed rectangles of the i-th and j-th polygons, A i and A j are the areas of the i-th and j-th polygons respectively; Define the following constraints: The minimum distance between polygons is not less than a preset value d min ; The maximum area of ​​the open space does not exceed a preset value S max ; The area of ​​each polygon after updating the coordinates is equal to the original area; The objective function and constraints are combined into an optimization problem, and nonlinear programming is used to solve the optimization problem and obtain the optimal solution; finally, the coordinates of the relevant points and lines are updated through the calculated optimal solution point.