A method and device for mining key network nodes based on multi-type node features

Abstract

The present invention relates to a method and device for mining key nodes in a network based on multi-class node features. The method comprises: calculating the degree of each node according to the connection status of each node in the network; further calculating the H-index of each node according to the degree of each node; calculating the K-shell index of each node according to the K-shell decomposition method; calculating the R-order neighborhood of each node according to the breadth-first search method; and calculating the importance score of each node according to the degree, H-index, K-shell index and R-order neighborhood of each node in the network. The present invention effectively considers multi-class node features, and thus can more comprehensively reflect the importance of nodes in the network, more accurately mine key nodes in the network, and reduce the misjudgment rate. In addition, the present invention only considers the R-order neighborhood of each node in the network, which can reduce time complexity on the one hand and adjust the size of R according to actual conditions on the other hand, making the method more flexible and adaptable.

Description

Technical Field

[0001] The present invention relates to the field of network information mining, and in particular to a method and device for mining key network nodes based on multi-type node features. Background Art

[0002] Due to the heterogeneity of complex networks, different nodes can play vastly different roles in their structure and function. Key nodes are those special nodes that can significantly influence the structure and function of a network. Key nodes are generally few in number, but their influence is immense, rapidly spreading across the majority of nodes in the network. Superstars can significantly accelerate the spread of information, superspreaders of infectious diseases can significantly amplify the disease's scale, a deliberate attack on a few critical servers can cause the collapse of the entire network, and failures in a small number of key power stations can lead to catastrophic power outages. Therefore, discovering key nodes in a network is of paramount importance, and the key to solving these problems lies in designing effective key node discovery methods.

[0003] Currently, most key node mining methods utilize only structural information from the network and are primarily categorized into neighbor-based centrality and path-based centrality. Neighbor-based centrality methods include degree centrality, the H-index, and the K-shell decomposition method, while path-based centrality methods include betweenness centrality and closeness centrality. However, none of these classic key node mining methods can comprehensively and accurately assess the importance of nodes in the network. Summary of the Invention

[0004] The present invention provides a method and device for mining key nodes in a network based on multi-class node features, the purpose of which is to solve the problem of incomplete evaluation and low accuracy of key nodes in classical methods by effectively combining multi-class node features.

[0005] Technical solution:

[0006] A method for mining key network nodes based on multi-type node features, characterized in that the method comprises:

[0007] Collect network data, abstract the entities in the network into nodes, number the nodes, and obtain the node set V = {V1, V2, ..., V N}, where N is the total number of nodes, and the connection status of each node with other nodes is recorded;

[0008] Obtaining an adjacency matrix of the network according to the connection status of each node;

[0009] Calculating the degree of each node based on the connectivity of each node in the network. The degree of the node is used to characterize the number of neighbors of the node.

[0010] Calculate the H index of each node based on the degree of the node, and the H index of the node is used to characterize the quality of the node's neighbors;

[0011] The K-shell index of each node is calculated according to the K-shell decomposition method. The K-shell index of the node is used to characterize whether the node is located in the core position of the network;

[0012] The R-order neighborhood of each node is calculated according to the breadth-first search method. The R-order neighborhood of the node includes all nodes whose distance to the node is not greater than R, which is used to characterize the distance between nodes.

[0013] The importance score of each node in the network is calculated according to the degree, H index, K shell index and R order neighborhood of each node, thereby obtaining the key nodes in the network.

[0014] Furthermore, the adjacency matrix of the network is: if there is an edge between node i and node j, the element corresponding to the i-th row and j-th column of the adjacency matrix is 1, otherwise it is 0.

[0015] Furthermore, the degree of a node in the network is equal to the number of direct edges connecting the node.

[0016] Furthermore, the calculation of the H-index of the node specifically includes:

[0017] Input: the degree of each node in the network node set V;

[0018] Steps: Count the degrees of neighboring nodes corresponding to the node, set the initial value of the node's H index to h=1, and repeat the following process:

[0019] 1) Determine whether the node has h+1 neighbor nodes with a degree not less than h+1;

[0020] 2) If it exists, then h = h + 1;

[0021] 3) If it does not exist, the H index is h;

[0022] Output: H index of each node in the network node set V.

[0023] Furthermore, the calculation of the K-shell index specifically includes:

[0024] Input: adjacency matrix of the network;

[0025] Steps: First, remove the nodes with degree 1 in the network. At this time, new nodes with degree 1 may appear in the remaining network. Then remove them again until there are no more nodes with degree 1 in the remaining network. The K-shell index of all the nodes removed in this process is 1; then remove the nodes with degree 2 in the remaining network in the same way. The K-shell index of all the nodes removed in this process is 2; repeat this process until there are no more nodes in the network;

[0026] Output: K-shell index of each node in the network node set V.

[0027] Furthermore, the calculation of the R-order neighborhood specifically includes:

[0028] Input: adjacency matrix of the network, neighborhood size R;

[0029] Steps: First search all nodes with a distance of 1 from the node, then search all nodes with a distance of 2 from the node, and end after searching all nodes with a distance of R from the node;

[0030] Output: the R-order neighborhood of each node in the network node set V.

[0031] Furthermore, the importance score formula of each node is:

[0032]

[0033] Among them, k(i) is the degree of node i, h(i) is the H index of node i, k s (i) is the K-shell index of node i, and d(i, j) is the distance between nodes i and j.

[0034] A method and device for mining key network nodes based on multi-class node features, comprising a collection module, an adjacency matrix generation module, a node degree calculation module, a node H index calculation module, a node K shell index calculation module, a node R-order neighborhood calculation module and a node importance score calculation module connected in sequence, wherein the collection module collects network data, abstracts entities in the network into nodes, numbers the nodes, and obtains a node set V = {V1, V2, ..., V N}, where N is the total number of nodes, recording the connection status of each node with other nodes; an adjacency matrix generation module, which obtains the adjacency matrix of the network according to the connection status of each node; a node degree calculation module, which calculates the degree of each node according to the connection status of each node in the network, and the degree of the node is used to characterize the number of neighbors of the node; a node H-index calculation module, which calculates the H-index of each node according to the degree of the node, and the H-index of the node is used to characterize the quality of the node's neighbors; a node K-shell index calculation module, which calculates the K-shell index of each node according to the K-shell decomposition method, and the K-shell index of the node is used to characterize whether the node is located in the core position of the network; a node R-order neighborhood calculation module, which calculates the R-order neighborhood of each node according to the breadth-first search method, and the R-order neighborhood of the node includes all nodes with a distance from the node not greater than R, which is used to characterize the distance between nodes; a node importance score calculation module, which calculates the importance score of each node according to the degree, H-index, K-shell index and R-order neighborhood of each node in the network, and then obtains the key nodes in the network.

[0035] An electronic device comprising a processor, a memory, an input device, and an output device connected via a bus, wherein the processor is used to execute a computer program; the memory is used to store the computer program; wherein the computer program is used to be stored by the memory and loaded by the processor to execute the method for mining key network nodes based on multi-type node features as described in any one of claims 1-7.

[0036] A computer-readable storage medium stores a computer program, which, when executed by a processor, implements the method for mining key network nodes based on multi-type node features as described in any one of claims 1 to 7.

[0037] Beneficial effects: The present invention effectively considers the characteristics of multiple types of nodes, and thus can more comprehensively reflect the importance of nodes in the network, more accurately mine key nodes in the network, and reduce the misjudgment rate; and the present invention only considers the R-order neighborhood of each node in the network, which can reduce the time complexity on the one hand, and on the other hand, can adjust the size of R according to actual conditions, making the method more flexible and adaptable. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] Figure 1 This is a flow chart of the identification method in an embodiment of the present invention;

[0039] Figure 2 A network node connection diagram in an embodiment of the present invention;

[0040] Figure 3 This is a structural diagram of an identification device in an embodiment of the present invention;

[0041] Figure 4 Schematic diagram of the structure of an electronic device in an embodiment of the present invention. DETAILED DESCRIPTION

[0042] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention and not all of the embodiments. All other embodiments obtained by ordinary technicians in this field based on the embodiments of the present invention without making any creative efforts shall fall within the scope of protection of the present invention.

[0043] To identify key nodes in a network, in addition to considering the number of directly influenced neighboring nodes (described by the node's degree), the quality of its neighboring nodes (described by the node's H-index) and whether the node's network position (described by the node's K-shell index) is close to the network core should also be taken into account. Furthermore, the objective fact that the mutual influence between nodes gradually weakens as the distance between them increases (described by the node's R-order neighborhood) should also be considered. Therefore, key node mining methods need to consider the role of all of these factors simultaneously. However, most classic methods only consider a single factor, making it impossible to comprehensively and accurately evaluate the importance of nodes in the network using existing methods.

[0044] The present invention proposes a method for mining key network nodes based on multi-type node features. The degree is used to characterize the number of neighbor nodes of a node, the H index is used to characterize the quality of the neighbor nodes of a node, the K-shell index is used to characterize the network position of the node, and the R-order neighborhood is used to characterize the influence of the distance between nodes. On this basis, the four are effectively combined to achieve the purpose of comprehensively evaluating the importance of nodes.

[0045] The present invention does not limit the network type involved. Figure 1 FIG. 1 is a method for mining key network nodes based on multi-type node features according to an embodiment of the present invention, comprising:

[0046] S100 collects network data, abstracts the entities in the network into nodes, numbers the nodes, and obtains a node set V = {V1, V2, ..., V N}, where N is the total number of nodes, and the connection status of each node with other nodes is recorded.

[0047] Specifically, if there is a direct association between the two collected entities, there is a direct edge between the corresponding two nodes; if there is no direct association between the two entities, there is no direct edge between the corresponding two nodes.

[0048] S110 obtains the adjacency matrix of the network according to the connection status of each node.

[0049] Specifically, if there is an edge between node i and node j, the element corresponding to the i-th row and j-th column of the adjacency matrix is 1, otherwise it is 0.

[0050] S120 calculates the degree of each node according to the connection status of each node in the network. The degree of the node is used to characterize the number of neighbors of the node.

[0051] Specifically, the degree of a node in a network is equal to the number of direct edges connecting the node.

[0052] S130 calculates the H-index of each node according to the degree of the node. The H-index of the node is used to characterize the quality of the node's neighbors.

[0053] The H-index of each node is calculated based on the degree of the node, specifically including:

[0054] Input: the degree of each node in the network node set V;

[0055] Steps: Count the degrees of neighboring nodes corresponding to the node, set the initial value of the node's H index to h=1, and repeat the following process:

[0056] 1) Determine whether the node has h+1 neighbor nodes with a degree not less than h+1;

[0057] 2) If it exists, then h = h + 1;

[0058] 3) If it does not exist, the H index is h;

[0059] Output: H index of each node in the network node set V.

[0060] S140 calculates the K-shell index of each node according to the K-shell decomposition method. The K-shell index of the node is used to characterize whether the node is located in the core position of the network.

[0061] The K-shell index of each node is calculated according to the K-shell decomposition method, specifically including:

[0062] Input: adjacency matrix of the network;

[0063] Steps: First, remove the nodes with degree 1 in the network. At this time, new nodes with degree 1 may appear in the remaining network. Then remove them again until there are no more nodes with degree 1 in the remaining network. The K-shell index of all the nodes removed in this process is 1; then remove the nodes with degree 2 in the remaining network in the same way. The K-shell index of all the nodes removed in this process is 2; repeat this process until there are no more nodes in the network;

[0064] Output: K-shell index of each node in the network node set V.

[0065] S150 calculates the R-order neighborhood of each node according to the breadth-first search method. The R-order neighborhood of the node includes all nodes whose distance to the node is not greater than R, and is used to characterize the distance feature between nodes.

[0066] The calculation of the R-order neighborhood of each node according to the breadth-first search method specifically includes:

[0067] Input: the adjacency matrix of the network, the neighborhood size R;

[0068] Steps: First search all nodes with a distance of 1 from the node, then search all nodes with a distance of 2 from the node, and end after searching all nodes with a distance of R from the node;

[0069] Output: the R-order neighborhood of each node in the network node set V.

[0070] S160 calculates the importance score of each node in the network according to the degree, H index, K shell index and R-order neighborhood of each node, and then obtains the key nodes in the network.

[0071] Specifically, the importance score formula of each node is:

[0072]

[0073] Among them, k(i) is the degree of node i, h(i) is the H index of node i, k s (i) is the K-shell index of node i, and d(i,j) is the distance between node i and node j. According to this formula, the higher the importance score of a node, the more important the node is.

[0074] It is worth noting that the method flow chart in the embodiment of the present invention is intended to more clearly illustrate the technical solution in the embodiment of the present invention, and does not constitute a limitation on the technical solution provided in the embodiment of the present invention. Moreover, the embodiment of the present invention is not limited to network applications. For other system structures and business applications, the technical solution provided in the embodiment of the present invention is also applicable to similar problems.

[0075] The following will be combined Figure 2 The example network in the figure describes the specific implementation of the present invention in detail.

[0076] Step 1: Since the example network has been given, data collection, node numbering, and recording the connection status of each node are omitted.

[0077] Step 2: Obtain the adjacency matrix of the network based on the connection status of each node in the network. The adjacency matrix corresponding to the network in this example can be expressed as:

[0078]

[0079] Step 3: Calculate the degree of each node. Take node 2 as an example. Node 2 has 5 edges, so k(2) = 5. The degrees of all nodes are shown in Table 1:

[0080] Table 1 Degree of each node

[0081]

[0082] Step 4: Calculate the H-index of each node. Taking node 2 as an example, the neighbor nodes of node 2 are 1, 3, 4, 5, and 6, and the corresponding degrees are 2, 4, 3, 4, and 1, respectively. Since node 2 has at most 3 neighbor nodes with a degree not less than 3, and there are no 4 neighbor nodes with a degree not less than 4, h(2) = 3. The H-index of all nodes is shown in Table 2:

[0083] Table 2 H index of each node

[0084]

[0085] Step 5: Calculate the K-shell index of each node. Remove the nodes (6, 8, 10) with degree 1 and their edges in the network. There is still node 9 with degree 1 in the network. Remove node 9 and its edges. There is still node 7 with degree 1 in the network. Remove node 7 and its edges. There are no more nodes with degree 1 in the network. Therefore, the K-shell index of nodes 6, 7, 8, 9, and 10 is 1. The network now contains five nodes (1, 2, 3, 4, and 5). Remove node 1 with degree 2 and its edges. There are no more nodes with degree 2 in the network. Therefore, the K-shell index of node 1 is 2. The network now has only four nodes (2, 3, 4, and 5). Remove the nodes (2, 3, 4, and 5) with degree 3 and their edges. There are no more nodes in the network. Therefore, the K-shell index of nodes 2, 3, 4, and 5 is 3. The K-shell indexes of all nodes are shown in Table 3:

[0086] Table 3 K-shell index of each node

[0087]

[0088] Step 6: Calculate the R-order neighborhood of each node. Taking the 2-order neighborhood as an example (i.e., R=2), the 2-order neighborhood of a node includes nodes with a distance of 1 from the node (i.e., 1-order neighbors) and nodes with a distance of 2 from the node (i.e., 2-order neighbors). The R-order neighborhoods of all nodes are shown in Table 4:

[0089] Table 4 R-order neighborhood of each node

[0090]

[0091] Step 7: Calculate the importance score of each node. The higher the score, the more important the node. Taking node 2 as an example, its first-order neighbors include nodes 1, 3, 4, 5, and 6, and its second-order neighbors include node 7. Substituting the degree, H index, and K-shell index of the above nodes into the node importance score formula, we can calculate S(2) = 451. The importance scores of all nodes are shown in Table 5:

[0092] Table 5 Importance score of each node

[0093]

[0094] According to the importance scores in Table 5, the final node importance ranking can be obtained, as shown in Table 6:

[0095] Table 6 Importance ranking of each node

[0096]

[0097] From Tables 1, 2, 3, and 5, it can be seen that, unlike the phenomenon that a large number of nodes have the same index value in terms of node degree, H index, and K-shell index, the method proposed in the present invention can distinguish the differences between nodes at a deeper level. Taking nodes 6, 8, and 10 as examples, although the three are leaf nodes, their network positions are different. Obviously, node 6 is closer to the core of the network, while node 10 is at the edge of the network. Therefore, node 6 is more important than node 10. Since the method proposed in the present invention effectively considers the characteristics of multiple types of nodes, it can more comprehensively reflect the importance of nodes in the network, more accurately mine key nodes in the network, and reduce the misjudgment rate. In addition, the present invention only considers the R-order neighborhood of each node in the network. On the one hand, it can reduce the time complexity, and on the other hand, it can adjust the size of R according to actual conditions, making the method more flexible and more adaptable.

[0098] Based on the above embodiments, see Figure 3 As shown, in an embodiment of the present invention, the device for mining key network nodes based on multiple types of node features specifically includes: a collection module 20, an adjacency matrix generation module 21, a node degree calculation module 22, a node H index calculation module 23, a node K shell index calculation module 24, a node R-order neighborhood calculation module 25 and a node importance score calculation module 26, which are connected in sequence.

[0099] The acquisition module 20 collects network data, abstracts the entities in the network into nodes, numbers the nodes, and obtains a node set V = {V1, V2, ..., V N}, where N is the total number of nodes, and the connection status of each node with other nodes is recorded;

[0100] The adjacency matrix generation module 21 obtains the adjacency matrix of the network according to the connection status of each node;

[0101] A node degree calculation module 22 calculates the degree of each node based on the connectivity of each node in the network. The node degree is used to characterize the number of neighbors of the node.

[0102] A node H-index calculation module 23 calculates the H-index of each node based on the degree of the node. The H-index of the node is used to characterize the quality of the node's neighbors.

[0103] The node K-shell index calculation module 24 calculates the K-shell index of each node according to the K-shell decomposition method. The K-shell index of the node is used to characterize whether the node is located in the core position of the network;

[0104] The node R-order neighborhood calculation module 25 calculates the R-order neighborhood of each node according to the breadth-first search method. The R-order neighborhood of a node includes all nodes whose distance to the node is not greater than R, and is used to characterize the distance between nodes.

[0105] The node importance score calculation module 26 calculates the importance score of each node in the network according to the degree, H index, K shell index and R-order neighborhood of each node, and then obtains the key nodes in the network.

[0106] Further, including:

[0107] In the adjacency matrix generation module 21, the adjacency matrix of the network is obtained according to the connection status of each node, specifically: if there is an edge between node i and node j, the element corresponding to the i-th row and j-th column of the adjacency matrix is 1, otherwise it is 0.

[0108] Further, including:

[0109] In the node degree calculation module 22, the degree of each node is calculated according to the connection status of each node in the network. Specifically, the degree of a node in the network is equal to the number of directly connected edges of the node.

[0110] Further, including:

[0111] In the node H-index calculation module 23, the H-index of each node is calculated according to the degree of the node, specifically including:

[0112] Input: the degree of each node in the network node set V;

[0113] Steps: Count the degrees of neighboring nodes corresponding to the node, set the initial value of the node's H index to h=1, and repeat the following process:

[0114] 1) Determine whether the node has h+1 neighbor nodes with a degree not less than h+1;

[0115] 2) If it exists, then h = h + 1;

[0116] 3) If it does not exist, the H index is h;

[0117] Output: H index of each node in the network node set V.

[0118] Further, including:

[0119] In the node K-shell index calculation module 24, the K-shell index of each node is calculated according to the K-shell decomposition method, specifically including:

[0120] Input: adjacency matrix of the network;

[0121] Steps: First, remove the nodes with degree 1 in the network. At this time, new nodes with degree 1 may appear in the remaining network. Then remove them again until there are no more nodes with degree 1 in the remaining network. The K-shell index of all the nodes removed in this process is 1; then remove the nodes with degree 2 in the remaining network in the same way. The K-shell index of all the nodes removed in this process is 2; repeat this process until there are no more nodes in the network;

[0122] Output: K-shell index of each node in the network node set V.

[0123] Further, including:

[0124] In the node R-order neighborhood calculation module 25, the calculation of the R-order neighborhood of each node according to the breadth-first search method specifically includes:

[0125] Input: the adjacency matrix of the network, the neighborhood size R;

[0126] Steps: First search all nodes with a distance of 1 from the node, then search all nodes with a distance of 2 from the node, and end after searching all nodes with a distance of R from the node;

[0127] Output: the R-order neighborhood of each node in the network node set V.

[0128] Further, including:

[0129] In the node importance score calculation module 26, the importance score formula of each node is:

[0130]

[0131] Among them, k(i) is the degree of node i, h(i) is the H index of node i, k s (i) is the K-shell index of node i, and d(i, j) is the distance between nodes i and j.

[0132] See Figure 4 FIG. 1 is a schematic diagram of the structure of an electronic device in an embodiment of the present invention.

[0133] An embodiment of the present invention provides an electronic device that may include a processor 310 (Center Processing Unit, CPU), memory 320, input device 330, and output device 340. These four devices are connected via a bus. Input device 330 is responsible for reading data, then passing the data via the bus to processor 310 and memory 320 for processing, and finally returning the processing results to output device 340 via the bus. Input device 330 may include a keyboard, mouse, touch screen, etc. Output device 340 may include a display device such as a liquid crystal display (LCD) or cathode ray tube (CRT).

[0134] The memory 320 may include a read-only memory (ROM) and a random access memory (RAM), and provides program instructions and data stored in the memory 320 to the processor 310. In an embodiment of the present invention, the memory 320 may be used to store the program of the geographic information query method.

[0135] The processor 310 calls the program instructions stored in the memory 320, and the processor 310 is used to execute the steps of any of the above methods for mining key network nodes based on multi-type node features according to the obtained program instructions.

[0136] Based on the above embodiments, in an embodiment of the present invention, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the method of mining key network nodes based on multiple types of node features in any of the above method embodiments is implemented.

[0137] It will be understood by those skilled in the art that embodiments of the present invention may be provided as methods, systems, or computer program products. Thus, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware. Furthermore, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to magnetic disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0138] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowcharts and / or block diagrams, as well as combinations of processes and / or blocks in the flowcharts and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowcharts and / or block diagrams. Figure 1 a process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0139] These computer program instructions may also be stored in a computer readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 a process or multiple processes and / or boxes Figure 1 The function specified in one or more boxes.

[0140] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operational steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing the instructions executed on the computer or other programmable device for implementing the process. Figure 1 a process or multiple processes and / or boxes Figure 1 The steps for the function specified in one or more boxes.

[0141] Although the preferred embodiments of the present invention have been described, those skilled in the art may make additional changes and modifications to these embodiments once they have learned the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications that fall within the scope of the present invention.

[0142] Obviously, those skilled in the art may make various changes and modifications to the embodiments of the present invention without departing from the spirit and scope of the embodiments of the present invention. Thus, if such changes and modifications of the embodiments of the present invention fall within the scope of the claims and their equivalents, the present invention is intended to include such changes and modifications.

Claims

1. A method for mining key network nodes based on multi-class node features, characterized in that: The method includes: Collect network data, abstract the entities in the network into nodes, number the nodes, and obtain the node set V = {V1, V2, ..., V N }, where N is the total number of nodes, and the connection status of each node with other nodes is recorded; Obtaining an adjacency matrix of the network according to the connection status of each node; Calculating the degree of each node based on the connectivity of each node in the network. The degree of the node is used to characterize the number of neighbors of the node. Calculate the H index of each node based on the degree of the node, and the H index of the node is used to characterize the quality of the node's neighbors; The K-shell index of each node is calculated according to the K-shell decomposition method. The K-shell index of the node is used to characterize whether the node is located in the core position of the network; The R-order neighborhood of each node is calculated according to the breadth-first search method. The R-order neighborhood of the node includes all nodes whose distance to the node is not greater than R, which is used to characterize the distance between nodes. Calculate the importance score of each node in the network according to the degree, H index, K shell index and R order neighborhood of each node, and then obtain the key nodes in the network; The importance score formula of each node is: Among them, k(i) is the degree of node i, h(i) is the H index of node i, k s (i) is the K-shell index of node i, and d(i, j) is the distance between nodes i and j.

2. The method for mining key network nodes based on multi-type node features according to claim 1 is characterized in that: The adjacency matrix of the network is: if there is an edge between node i and node j, the element corresponding to the i-th row and j-th column of the adjacency matrix is 1, otherwise it is 0.

3. The method for mining key network nodes based on multi-type node features according to claim 1 is characterized in that: The degree of a node in the network is equal to the number of direct edges connecting the node.

4. The method for mining key network nodes based on multi-type node features according to claim 1 is characterized in that: The calculation of the H-index of the node specifically includes: Input: the degree of each node in the network node set V; Steps: Count the degrees of neighboring nodes corresponding to the node, set the initial value of the node's H index to h=1, and repeat the following process: 1) Determine whether the node has h+1 neighbor nodes with a degree not less than h+1; 2) If it exists, then h = h + 1; 3) If it does not exist, the H index is h; Output: H index of each node in the network node set V.

5. The method for mining key network nodes based on multi-type node features according to claim 1 is characterized in that: The calculation of the K-shell index specifically includes: Input: adjacency matrix of the network; Steps: First, remove the nodes with degree 1 in the network. At this time, new nodes with degree 1 may appear in the remaining network. Then remove them again until there are no more nodes with degree 1 in the remaining network. The K-shell index of all the nodes removed in this process is 1; then remove the nodes with degree 2 in the remaining network in the same way. The K-shell index of all the nodes removed in this process is 2; repeat this process until there are no more nodes in the network; Output: K-shell index of each node in the network node set V.

6. The method for mining key network nodes based on multi-type node features according to claim 1, characterized in that: The calculation of the R-order neighborhood specifically includes: Input: adjacency matrix of the network, neighborhood size R; Steps: First search all nodes with a distance of 1 from the node, then search all nodes with a distance of 2 from the node, and end after searching all nodes with a distance of R from the node; Output: the R-order neighborhood of each node in the network node set V.

7. A device for the method of mining key network nodes based on multiple types of node features as claimed in claim 1, characterized in that: It includes a collection module (20), an adjacency matrix generation module (21), a node degree calculation module (22), a node H index calculation module (23), a node K shell index calculation module (24), a node R-order neighborhood calculation module (25) and a node importance score calculation module (26) which are connected in sequence. The acquisition module (20) collects network data, abstracts the entities in the network into nodes, numbers the nodes, and obtains a node set V = {V1, V2, ..., V N }, where N is the total number of nodes, and the connection status of each node with other nodes is recorded; An adjacency matrix generation module (21) obtains the adjacency matrix of the network according to the connection status of each node; A node degree calculation module (22) calculates the degree of each node according to the connection status of each node in the network, wherein the degree of the node is used to characterize the number of neighbors of the node; A node H-index calculation module (23) calculates the H-index of each node according to the degree of the node, and the H-index of the node is used to characterize the quality of the node's neighbors; A node K-shell index calculation module (24) calculates the K-shell index of each node according to the K-shell decomposition method, wherein the K-shell index of the node is used to characterize whether the node is located in the core position of the network; A node R-order neighborhood calculation module (25) calculates the R-order neighborhood of each node according to a breadth-first search method. The R-order neighborhood of a node includes all nodes whose distance to the node is not greater than R, and is used to characterize the distance between nodes. The node importance score calculation module (26) calculates the importance score of each node according to the degree, H index, K shell index and R order neighborhood of each node in the network, and then obtains the key nodes in the network.

8. An electronic device, characterized in that: The invention comprises a processor (310), a memory (320), an input device (330) and an output device (340) connected via a bus, wherein the processor (310) is used to execute a computer program; the memory (320) is used to store the computer program; wherein the computer program is used to be stored by the memory (320) and loaded by the processor (310) to execute the method for mining key network nodes based on multi-type node features as described in any one of claims 1 to 7.

9. A computer-readable storage medium, characterized in that The storage medium stores a computer program, which, when executed by the processor (310), implements the method for mining key network nodes based on multi-type node features as described in any one of claims 1 to 7.

Images