CN109995663B - Network information propagation method based on length constraint and no repeated path - Google Patents

Network information propagation method based on length constraint and no repeated path Download PDF

Info

Publication number
CN109995663B
CN109995663B CN201910224582.4A CN201910224582A CN109995663B CN 109995663 B CN109995663 B CN 109995663B CN 201910224582 A CN201910224582 A CN 201910224582A CN 109995663 B CN109995663 B CN 109995663B
Authority
CN
China
Prior art keywords
node
network
nodes
optimal path
path
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Active
Application number
CN201910224582.4A
Other languages
Chinese (zh)
Other versions
CN109995663A (en
Inventor
吴建设
周超杰
丁振林
丁琪琪
管铭
孙文静
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Xidian University
Original Assignee
Xidian University
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Xidian University filed Critical Xidian University
Priority to CN201910224582.4A priority Critical patent/CN109995663B/en
Publication of CN109995663A publication Critical patent/CN109995663A/en
Application granted granted Critical
Publication of CN109995663B publication Critical patent/CN109995663B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
    • H04L41/14—Network analysis or design
    • H04L41/145—Network analysis or design involving simulating, designing, planning or modelling of a network
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L45/00—Routing or path finding of packets in data switching networks
    • H04L45/02—Topology update or discovery
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L45/00—Routing or path finding of packets in data switching networks
    • H04L45/12—Shortest path evaluation

Landscapes

  • Engineering & Computer Science (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Signal Processing (AREA)
  • Data Exchanges In Wide-Area Networks (AREA)

Abstract

A network information propagation method based on length constraint and no repeated path mainly solves the problem of trust propagation of users in a social network. The method comprises the following steps: 1. establishing a trust network graph model; 2. determining an intermediate network node; 3. selecting an optimal path; 4. calculating a network node information value; 5. propagating the network message; 6. the untrusted network information is discarded. According to the invention, by establishing the trust network graph model and selecting the optimal path, the structural attribute of the trust network is effectively maintained, and the availability of the trust network is improved. The invention uses the total number of the non-repeated paths of all the path lengths from the optimal path starting node to the terminal node to calculate the network information acceptance degree from the starting node to the terminal node, thereby effectively improving the accuracy of the network information between the nodes while ensuring the integrity of the trust network.

Description

Network information propagation method based on length constraint and no repeated path
Technical Field
The invention belongs to the technical field of computers, and further relates to a network information propagation method based on length constraint and no repeated path in the technical field of data mining. The method can be used for spreading trust information such as characters and evaluation among different nodes in complex networks such as microblog, social contact and e-commerce.
Background
The social network, the e-commerce network and the like are typical complex networks, each person or each individual is abstracted into a network node in the complex network, the relationship between people is abstracted into connecting edges between corresponding nodes in the network, the association degree between the nodes can be abstracted into the weight of the corresponding edges in the network, and therefore a complex network model is formed, and the trust information of adjacent nodes in the network can be represented by the weight between the nodes. The information propagation model in the complex network not only can make clear the process of information propagation, but also can predict the propagation path and the propagation situation of the information, and macroscopically and microscopically grasp the characteristics of information propagation, thereby providing a theoretical basis for the research of information propagation. At present, the trust propagation in the complex network is mainly a trust propagation model based on matrix power and a trust propagation model based on paths among nodes.
The patent document "model and method for propagating social network-oriented multi-information and multi-dimensional network information" (patent application No. 201711380291.1, application publication No. CN 108230170a) applied by Chongqing post and telecommunications university discloses a method for propagating social network-oriented multi-information and multi-dimensional network information. According to the method, the social network data are acquired and preprocessed to construct the multidimensional space network, influence factors are introduced to express the interaction and strength among different information, and information propagation of the multidimensional space network is determined by a cosine similarity method. The method has the disadvantages that because the method introduces additional influence factors, the uncertainty of the relationship between the nodes is increased, and the trust relationship between the nodes cannot be accurately described.
Zheng zhou university discloses a social network trust degree calculation method based on propagation path privacy disclosure risk in the patent document "social network trust degree calculation method based on propagation path privacy disclosure risk" (patent application No. 201810654835.7, application publication No. CN 108900409 a) filed by zheng zhou university. The method comprises the steps of establishing a blacklist for each user in the social network, limiting information interaction between the users in the blacklist and the users in the network, and calculating the trust between the users by deleting local network connection edges randomly in a social network platform, predicting an information propagation path and privacy disclosure risk probability. The method has the defects that the local network connection edges are deleted randomly in the social network platform, so that the topological structure of the network is damaged to a greater extent, and the deleted network has low availability.
Disclosure of Invention
The invention aims to provide a network information propagation method based on length constraint and no repeated path to solve the problem of information propagation reliability in a network aiming at the defects of the prior art.
The idea of realizing the purpose of the invention is that firstly, a trust network graph model is used, in the process of searching the optimal network path from an initial node to a terminal node, the influence of the weight value between path nodes is considered, the structural attribute of the trust network is ensured not to be changed, the subsequent nodes of the path are determined by using a path cost estimation value method, the length doubling operation is introduced, the total number of repeated paths with length constraint between the initial node and the terminal node is obtained, and the authenticity of the information transmitted between the nodes is judged by using the acceptance of the network information.
In order to achieve the above object, the method comprises the following steps:
(1) establishing a trust network graph model;
(1a) reading a data set at least containing 300 network nodes;
(1b) abstracting the data of the read data set into a directed graph;
(1c) generating an adjacency matrix corresponding to the directed graph by adopting a network adjacency matrix generation method;
(2) determining an intermediate network node:
(2a) respectively establishing an empty stack for storing a subsequent node, an empty set for storing a terminal node and an intersection for storing a network node;
(2b) randomly selecting an unselected network node from the data set of the network node as an initial node;
(2c) storing all network nodes which have connecting edges with the initial node and have never been visited as successor nodes in a stack;
(2d) popping a successor node from the stack;
(2e) judging whether the number of edges on a path from the current starting node to the current popped successor node is less than or equal to a length constraint value, if so, storing the current popped successor node as a terminal node into a set, and executing the step (2f), otherwise, executing the step (2c) by taking the current popped successor node as the starting node;
(2f) judging whether the stack is empty, if so, obtaining a network set, and executing the step (2g), otherwise, executing the step (2 d);
(2g) judging whether two sets are obtained or not, if so, executing the step (2h), otherwise, executing the step (2 a);
(2h) judging whether network nodes with the same structure exist in the two sets, if so, executing the step (2i), otherwise, adding 1 to the length constraint value and executing the step (2 a);
(2i) putting all network nodes in any one set corresponding to all network nodes with the same structure into an intersection, and randomly selecting one network node from the intersection as an intermediate network node;
(2j) taking the current starting node as a target node;
(3) selecting an optimal path:
(3a) selecting unselected network nodes from the intersection as initial nodes of the optimal path, and using the target node as a terminal node of the optimal path;
(3b) establishing an empty set for storing candidate successor nodes, establishing a linked list for storing optimal path nodes, and establishing a set for storing paths;
(3c) taking a network node which has a connecting edge with the initial node of the optimal path and has never been visited as a candidate successor node, and storing the candidate successor node in a set;
(3d) selecting an unselected candidate successor node from the set;
(3e) calculating the total weight value of the path from the initial node to the selected subsequent node by using a total weight value calculation formula;
(3f) and calculating the cost estimation value of the optimal path from the selected successor node to the terminal node of the optimal path according to the following formula:
f＝g(·)+h(·)
wherein, f represents the cost estimation value from the initial node of the optimal path to the terminal node of the optimal path via the selected successor node, g (-) represents the cost value from the initial node of the optimal path to the selected successor node, and h (-) represents the optimal path cost estimation value from the currently selected successor node to the terminal node of the optimal path;
(3g) judging whether all the nodes in the candidate node set are selected, if so, executing the step (3g), otherwise, executing the step (3 d);
(3h) selecting the minimum value from the initial node to the terminal node cost estimation value from the set, and storing a subsequent node corresponding to the minimum value as an intermediate node of the optimal path into a linked list;
(3i) judging whether the intermediate node of the optimal path stored in the linked list is a terminal node, if so, putting the linked list into a set of storage paths and then executing the step (3j), otherwise, executing the step (3c) after taking the current candidate subsequent node as the initial node of the optimal path;
(3j) judging whether unselected nodes exist in the intersection, if so, executing the step (3a), otherwise, obtaining a storage path set, and executing the step (4);
(4) calculating a network node information value:
(4a) calculating the network information acceptance from the initial node of the optimal path to the terminal node according to the following formula:
Figure GDA0002330300920000041
wherein gamma represents the network information acceptance from the initial node of the optimal path to the terminal node, α represents the network information propagation credibility parameter, and ZNRepresents the total number of all paths with the length of N in the storage path set, sigma represents the summation operation, S represents a subset of the path node set,
Figure GDA0002330300920000043
denotes an empty set, ZN(S) represents the total number of paths through all path nodes in the subset S;
(4b) calculating the connection state ratio of the initial node of the optimal path according to the following formula:
Figure GDA0002330300920000042
wherein u represents the connection state ratio of the initial node of the optimal path, t represents the initial node of the optimal path, vnRepresents the nth node in the directed graph, e represents belonging to a symbol, V represents storing a set of candidate successor nodes,
Figure GDA0002330300920000044
starting node t representing optimal path and nth node v in directed graphnN represents the total number of all nodes in the directed graph;
(4c) judging whether the acceptance of the network information is larger than or equal to the ratio of the connection state, if so, executing the step (5), otherwise, executing the step (6);
(5) propagating the network message:
judging the network information from the initial node to the terminal node of the optimal path as information with high reliability, and carrying out a new round of information transmission;
(6) discarding untrusted network information:
and judging the network information transmitted from the starting node to the terminal node as unreliable, discarding the network information and terminating information propagation.
Compared with the prior art, the invention has the following advantages:
firstly, the invention uses the trust network graph model to select the optimal path from the starting node to the terminal node, thereby overcoming the problem that the deleted network has low availability caused by deleting local network connection edges at random of the social network and destroying the topological structure of the network to a greater extent in the prior art. The invention effectively ensures the structure attribute of the network and improves the usability of the social network while selecting the optimal path.
Secondly, the invention uses the total number of the non-repeated paths with the length of 2N from the optimal path starting node to the terminal node to calculate the network information acceptance from the starting node to the terminal node, thereby overcoming the problems that additional influence factors are introduced in the prior art, the uncertainty of the relationship between the nodes is increased, and the trust relationship between the nodes cannot be accurately described. The invention ensures the integrity of the trust network and effectively improves the reliability of the network information between the nodes.
Drawings
FIG. 1 is a flow chart of the present invention;
FIG. 2 is a diagram of the results of a simulation experiment of the present invention.
Detailed Description
The invention is further described below with reference to the accompanying drawings.
The specific steps implemented by the present invention are described in further detail with reference to fig. 1.
Step 1, establishing a trust network graph model.
A data set containing at least 300 network nodes is read in.
The data of the read-in data set is abstracted into a directed graph.
The directed graph is G1(Y, E), wherein Y represents directed graph G1Each network node represents a user in the network, E represents a directed graph G1And (3) the collection of edges of the middle network node, wherein each edge represents the connection relation between any two network nodes in the network.
And generating an adjacency matrix corresponding to the directed graph by adopting a network adjacency matrix generation method.
The method for generating the network adjacency matrix comprises the following steps:
step 1, generating an adjacency matrix with B rows and B columns, wherein the elements of the adjacency matrix are 0, B represents the total number of network nodes in the network, and the network nodes are numbered in sequence according to the reading sequence of the network nodes.
And 2, randomly selecting two unselected network nodes from the directed graph, judging whether a connecting edge exists between the two selected network nodes, if so, setting an element at a position corresponding to the selected node in the adjacent matrix to be 1, and otherwise, setting the element to be 0.
And 3, judging whether all the nodes in the digraph are selected completely, if so, finishing the adjacency matrix, and otherwise, executing the second step.
And 2, determining the intermediate network node.
And (2.1) respectively establishing an empty stack for storing the successor node, an empty set for storing the terminal node and an intersection for storing the network nodes.
The intersection refers to whether network nodes with the same structure exist in the two sets, if so, all the network nodes in any one set corresponding to all the network nodes with the same structure are placed into an empty set, and the set is called as the intersection.
(2.2) randomly selecting one unselected network node from the data set of network nodes as an initial node.
And (2.3) storing all network nodes which have connection edges with the initial node and have never visited as successor nodes in the stack.
(2.4) popping a successor node from the stack.
(2.5) judging whether the number of edges on the path from the current starting node to the current popped successor node is less than or equal to the length constraint value, if so, storing the current popped successor node as a terminal node into a set, and executing the step (2.6), otherwise, using the current popped successor node as the starting node, and executing the step (2.3).
The length constraint value is calculated by the following formula:
Figure GDA0002330300920000061
where Σ denotes the summation operation, v1Represents the current start node, v2Representing the current popped successor node, e representing belonging to the symbol, Y representing the directed graph G1The set of all the network nodes in the network,
Figure GDA0002330300920000062
denoted as current start node v1With the currently popped successor node v2If there are two continuous edges, the value is 1, there are no continuous edges, the value is 0, k represents the current starting pointThe length of the path from the node to the current terminal node.
And (2.6) judging whether the stack is empty, if so, obtaining a network set, and executing the step (2.7), otherwise, executing the step (2.4).
And (2.7) judging whether two sets are obtained or not, if so, executing the step (2.8), otherwise, executing the step (2.1).
And (2.8) judging whether the two sets have network nodes with the same structure, if so, executing the step (2.9), and otherwise, adding 1 to the length constraint value and executing the step (2.1).
And (2.9) putting all the network nodes in any one set corresponding to all the network nodes with the same structure into an intersection, and randomly selecting one network node from the intersection as an intermediate network node.
And (2.10) taking the current starting node as a target node.
And 3, selecting an optimal path.
And (3.1) selecting unselected network nodes from the intersection as initial nodes of the optimal path, and using the target node as a terminal node of the optimal path.
And (3.2) establishing an empty set for storing candidate successor nodes, establishing a linked list for storing optimal path nodes, and establishing a set for storing paths.
And (3.3) taking the network node which has a continuous edge with the initial node of the optimal path and has never visited as a candidate successor node, and storing the candidate successor node into the set.
And (3.4) selecting an unselected candidate subsequent node from the set.
The total weight value calculation formula is as follows:
Figure GDA0002330300920000071
where t represents the total weight value on the optimal path, Σ represents the summation operation, vm，vnRespectively representing the mth candidate successor node and the nth candidate successor node in the set, representing the symbol belonging to the element, and representing the storage candidate by VA set of successor nodes is selected,
Figure GDA0002330300920000072
representing candidate successor nodes vm，vnIn the middle state, if there is a continuous edge between them, the value is 1, and if there is no continuous edge, the value is 0.
And (3.5) calculating the total weight value of the path from the starting node to the selected subsequent node by using a total weight value calculation formula.
(3.6) calculating the cost estimation value of the optimal path from the selected successor node to the terminal node of the optimal path according to the following formula:
f＝g(·)+h(·)
wherein, f represents the cost estimation value from the starting node of the optimal path to the terminal node of the optimal path via the selected successor node, g (-) represents the cost value from the starting node of the optimal path to the selected successor node, and h (-) represents the optimal path cost estimation value from the currently selected successor node to the terminal node of the optimal path.
And (3.7) judging whether all the nodes in the candidate node set are selected, if so, executing the step (3.7), otherwise, executing the step (3.4).
(3.8) selecting the minimum value from the initial node to the terminal node cost estimation value from the set, and storing a subsequent node corresponding to the minimum value as an optimal path intermediate node into a linked list.
And (3.9) judging whether the intermediate node of the optimal path stored in the linked list is a terminal node, if so, putting the linked list into the set of the storage paths and then executing the step (3.10), and otherwise, executing the step (3.3) by taking the current candidate subsequent node as the initial node of the optimal path.
And (3.10) judging whether unselected nodes exist in the intersection, if so, executing the step (3.1), otherwise, obtaining a storage path set, and executing the step 4.
And 4, calculating the information value of the network node.
Calculating the network information acceptance from the initial node of the optimal path to the terminal node according to the following formula:
Figure GDA0002330300920000081
wherein gamma represents the network information acceptance from the initial node of the optimal path to the terminal node, α represents the network information propagation credibility parameter, and ZNRepresents the total number of all paths with the length of N in the storage path set, sigma represents the summation operation, S represents a subset of the path node set,
Figure GDA0002330300920000082
denotes an empty set, ZN(S) represents the total number of paths through all path nodes in the subset S.
The path node set refers to a set composed of nodes of all paths in the set of storage paths.
Calculating the connection state ratio of the initial node of the optimal path according to the following formula:
Figure GDA0002330300920000083
wherein u represents the connection state ratio of the initial node of the optimal path, σ represents the initial node of the optimal path, and GxDenotes the Xth node in the directed graph, e denotes belonging to a symbol, V denotes storing the set of candidate successor nodes,
Figure GDA0002330300920000084
starting node sigma representing optimal path and Xth node G in directed graphxN represents the total number of all nodes in the directed graph.
And (5) judging whether the acceptance of the network information is more than or equal to the ratio of the connection states, if so, executing the step 5, otherwise, executing the step 6.
And 5, propagating the network message.
And judging the network information from the initial node to the terminal node of the optimal path as information with high reliability, and carrying out a new round of information propagation.
And 6, discarding the untrusted network information.
And judging the network information transmitted from the starting node to the terminal node as unreliable, discarding the network information and terminating information propagation.
The effect of the invention can be further illustrated by the following simulation experiment:
1. simulation conditions are as follows:
the hardware test platform of the invention is: the processor is Intel (R) core (TM) i5-3470 CPU @3.20GHz, the memory is 4.0GB, and the hard disk is 500G; the software platform is as follows: microsoft windows 7 operating system and python 3.5.
2. Simulation content and simulation result analysis:
the simulation experiment of the invention adopts the method of the invention and the algorithm of the prior art, respectively on the Epinions network, randomly selects a part of nodes from the network to connect edges, calculates the acceptance of network information, takes the connection state ratio of the initial nodes as a threshold value, counts the total number of the predicted connecting edges lower than the threshold value as the total number of the predicted errors, takes the ratio of the total number of the predicted errors and the total number of the selected nodes to connect edges as the predicted error rate, takes the predicted error rate as the experiment result, respectively carries out 30 independent experiments, and takes the average value of the 30 experiment results as the simulation result of the experiment.
The prior art is a paper "de-routing algorithm and application" published by n.chen, j.wu, et al (wo thesis, 2017, pages 9-12).
The parameters of the simulation experiment of the present invention were set as follows, and the epipositions dataset included 50000 nodes and 487183 connected edges, each marked as trusted or untrusted. Trust is interpreted as the actual value 1 and distrust is interpreted as the actual value 0. 1500 edges are selected in the random selection process, the network information propagation credibility parameter is 0.3, and the experimental result is shown in fig. 2.
The abscissa in fig. 2 represents the value of the path length constraint and the ordinate represents the prediction error rate, wherein the curve marked with crosses represents the simulation result curve using the method of the present invention, and the curve marked with stars represents the simulation result curve using the prior art matrix power method. As can be seen from fig. 2, when the path constraint value of the abscissa increases, the method is significantly improved compared with the comparative method.
In summary, the invention adopts a network information propagation method based on length constraint and no repeated path, uses a trust network graph model, considers the influence of weight values between path nodes in the process of searching the optimal network path from the starting node to the terminal node, also ensures that the structural attribute of the trust network is not changed, determines the subsequent node of the path by using a path cost estimation value method, introduces length doubling operation to obtain the total number of the repeated paths without length constraint between the starting node and the terminal node, and judges the authenticity of the information propagated between the nodes by using the network information acceptance. The invention maintains the integrity of the network structure, removes the cyclic path between the nodes and obviously reduces the error rate.

Claims (5)

1. A network information transmission method based on length constraint and no repeated path is characterized in that an optimal path is selected and a network node information value is calculated, and the method specifically comprises the following steps:
(1) establishing a trust network graph model;
(1a) reading a data set at least containing 300 network nodes;
(1b) abstracting the data of the read data set into a directed graph;
(1c) generating an adjacency matrix corresponding to the directed graph by adopting a network adjacency matrix generation method;
(2) determining an intermediate network node:
(2a) respectively establishing an empty stack for storing a subsequent node, an empty set for storing a terminal node and an intersection for storing a network node;
(2b) randomly selecting an unselected network node from the data set of the network node as an initial node;
(2c) storing all network nodes which have connecting edges with the initial node and have never been visited as successor nodes in a stack;
(2d) popping a successor node from the stack;
(2e) judging whether the number of edges on a path from the current initial node to the current popped successor node is less than or equal to a length constraint value, if so, storing the current popped successor node as a terminal node into a set, and executing the step (2f), otherwise, executing the step (2c) by taking the current popped successor node as the initial node;
the length constraint value is calculated by the following formula:
Figure FDA0002330300910000011
where Σ denotes the summation operation, v1Represents the current start node, v2Representing the successor node currently popped up, e representing belonging to a symbol, Y representing the set of all network nodes in the directed graph,
Figure FDA0002330300910000012
denoted as current start node v1With the currently popped successor node v2If the two have connecting edges, the value is 1, and no connecting edge exists, the value is 0, and k represents the path length value from the current starting node to the current terminal node;
(2f) judging whether the stack is empty, if so, obtaining a network set, and executing the step (2g), otherwise, executing the step (2 d);
(2g) judging whether two sets are obtained or not, if so, executing the step (2h), otherwise, executing the step (2 a);
(2h) judging whether network nodes with the same structure exist in the two sets, if so, executing the step (2i), otherwise, adding 1 to the length constraint value and executing the step (2 a);
(2i) putting all network nodes in any one set corresponding to all network nodes with the same structure into an intersection, and randomly selecting one network node from the intersection as an intermediate network node;
(2j) taking the current starting node as a target node;
(3) selecting an optimal path:
(3a) selecting unselected network nodes from the intersection as initial nodes of the optimal path, and using the target node as a terminal node of the optimal path;
(3b) establishing an empty set for storing candidate successor nodes, establishing a linked list for storing optimal path nodes, and establishing a set for storing paths;
(3c) taking a network node which has a connecting edge with the initial node of the optimal path and has never been visited as a candidate successor node, and storing the candidate successor node in a set;
(3d) selecting an unselected candidate successor node from the set;
(3e) and calculating the total weight value of the path from the starting node to the selected successor node by using the following total weight value calculation formula:
Figure FDA0002330300910000021
wherein t represents the total weight value on the optimal path, V represents the set of candidate successor nodes, Vm，vnRespectively representing the mth candidate successor node and the nth candidate successor node in the set,
Figure FDA0002330300910000022
representing candidate successor nodes vmAnd vnThe value of the state between (1) and (0) if both have a connecting edge;
(3f) and calculating the cost estimation value of the optimal path from the selected successor node to the terminal node of the optimal path according to the following formula:
f＝g(·)+h(·)
wherein, f represents the cost estimation value from the initial node of the optimal path to the terminal node of the optimal path via the selected successor node, g (-) represents the cost value from the initial node of the optimal path to the selected successor node, and h (-) represents the optimal path cost estimation value from the currently selected successor node to the terminal node of the optimal path;
(3g) judging whether all the nodes in the candidate node set are selected, if so, executing the step (3h), otherwise, executing the step (3 d);
(3h) selecting the minimum value from the initial node to the terminal node cost estimation value from the set, and storing a subsequent node corresponding to the minimum value as an intermediate node of the optimal path into a linked list;
(3i) judging whether the intermediate node of the optimal path stored in the linked list is a terminal node, if so, putting the linked list into a set of storage paths and then executing the step (3j), otherwise, executing the step (3c) by taking the current candidate subsequent node as the initial node of the optimal path;
(3j) judging whether unselected nodes exist in the intersection, if so, executing the step (3a), otherwise, obtaining a storage path set, and executing the step (4);
(4) calculating a network node information value:
(4a) calculating the network information acceptance from the initial node of the optimal path to the terminal node according to the following formula:
Figure FDA0002330300910000031
wherein gamma represents the network information acceptance from the initial node of the optimal path to the terminal node, α represents the network information propagation credibility parameter, and ZNRepresenting the total number of non-duplicate paths of length N of all paths in the stored path set, S representing a subset of the path node set,
Figure FDA0002330300910000032
denotes an empty set, ZN(S) represents the total number of paths through all path nodes in the subset S;
(4b) calculating the connection state ratio of the initial node of the optimal path according to the following formula:
Figure FDA0002330300910000033
wherein u represents the connection state ratio of the initial node of the optimal path, σ represents the initial node of the optimal path, and GxRepresents the xth node in the directed graph,
Figure FDA0002330300910000034
starting node sigma representing optimal path and Xth node G in directed graphxN represents the total number of all nodes in the directed graph;
(4c) judging whether the acceptance of the network information is larger than or equal to the ratio of the connection state, if so, executing the step (5), otherwise, executing the step (6);
(5) propagating the network message:
judging the network information from the initial node to the terminal node of the optimal path as information with high reliability, and carrying out a new round of information transmission;
(6) discarding untrusted network information:
and judging the network information transmitted from the starting node to the terminal node as unreliable, discarding the network information and terminating information propagation.
2. The method according to claim 1, wherein the directed graph in step (1b) is G1(Y, E), wherein Y represents directed graph G1Each network node represents a user in the network, E represents a directed graph G1And (3) the collection of edges of the middle network node, wherein each edge represents the connection relation between any two network nodes in the network.
3. The method for network information propagation based on length constraint and no repeated path according to claim 1, wherein the step of the network adjacency matrix generation method in step (1c) is as follows:
firstly, generating an adjacent matrix of B rows and B columns with 0 elements, wherein B represents the total number of network nodes in the network, and numbering the network nodes in sequence according to the reading sequence of the network nodes;
second, two unselected network nodes are randomly selected from the directed graph, whether a connecting edge exists between the two selected network nodes is judged, if yes, an element of a position corresponding to the selected node in the adjacent matrix is set to be 1, and if not, the element is set to be 0;
and thirdly, judging whether all the nodes in the digraph are selected completely, if so, finishing the adjacency matrix, and otherwise, executing the second step.
4. The method according to claim 1, wherein the intersection in step (2a) is whether there are network nodes with the same structure in the two sets, and if yes, all network nodes in any one set corresponding to all network nodes with the same structure are put into an empty set, and the set is called an intersection.
5. The method according to claim 1, wherein the set of path nodes in step (4a) is a set of nodes of all paths in the set of storage paths.
CN201910224582.4A 2019-03-23 2019-03-23 Network information propagation method based on length constraint and no repeated path Active CN109995663B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201910224582.4A CN109995663B (en) 2019-03-23 2019-03-23 Network information propagation method based on length constraint and no repeated path

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201910224582.4A CN109995663B (en) 2019-03-23 2019-03-23 Network information propagation method based on length constraint and no repeated path

Publications (2)

Publication Number Publication Date
CN109995663A CN109995663A (en) 2019-07-09
CN109995663B true CN109995663B (en) 2020-04-07

Family

ID=67130964

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201910224582.4A Active CN109995663B (en) 2019-03-23 2019-03-23 Network information propagation method based on length constraint and no repeated path

Country Status (1)

Country Link
CN (1) CN109995663B (en)

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113329351B (en) * 2021-06-02 2022-08-30 合肥工业大学 Message transmission method and equipment thereof

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101764639A (en) * 2009-12-17 2010-06-30 哈尔滨工业大学 Multilayer satellite network stable grouping method based on five vector mathematical models
CN103995823A (en) * 2014-03-25 2014-08-20 南京邮电大学 Information recommending method based on social network
CN107133704A (en) * 2017-06-23 2017-09-05 南京师范大学 Follow the analogy method of the dynamic emergency evacuation of large-scale crowd of optimal-forgetting rules

Family Cites Families (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US7020394B2 (en) * 2001-08-17 2006-03-28 Quantum Bridge Communications, Inc. Method and apparatus for path selection and wavelength assignment in an optical network
US8290981B2 (en) * 2011-03-08 2012-10-16 Hon Hai Precision Industry Co., Ltd. Social network system and member searching and analyzing method in social network
CN103838820B (en) * 2013-12-24 2017-05-24 西安电子科技大学 Evolutionary multi-objective optimization community detection method based on affinity propagation
CN104102745B (en) * 2014-07-31 2017-12-29 上海交通大学 Complex network community method for digging based on Local Minimum side
CN108230170B (en) * 2017-12-20 2022-02-11 重庆邮电大学 Social network-oriented multi-information and multi-dimensional network information propagation model and method

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101764639A (en) * 2009-12-17 2010-06-30 哈尔滨工业大学 Multilayer satellite network stable grouping method based on five vector mathematical models
CN103995823A (en) * 2014-03-25 2014-08-20 南京邮电大学 Information recommending method based on social network
CN107133704A (en) * 2017-06-23 2017-09-05 南京师范大学 Follow the analogy method of the dynamic emergency evacuation of large-scale crowd of optimal-forgetting rules

Also Published As

Publication number Publication date
CN109995663A (en) 2019-07-09

Similar Documents

Publication Publication Date Title
Lian et al. An Intrusion Detection Method Based on Decision Tree‐Recursive Feature Elimination in Ensemble Learning
WO2019128529A1 (en) Url attack detection method and apparatus, and electronic device
CN108959370B (en) Community discovery method and device based on entity similarity in knowledge graph
CN113347210B (en) DNS tunnel detection method and device and electronic equipment
CN113015167B (en) Encrypted flow data detection method, system, electronic device and storage medium
CN109218304B (en) Network risk blocking method based on attack graph and co-evolution
CN110233833B (en) Message sending method and system supporting privacy protection of social network users
CN110222790B (en) User identity identification method and device and server
EP4351087A1 (en) Traffic detection method and apparatus, and device and readable storage medium
Elekar Combination of data mining techniques for intrusion detection system
CN109995663B (en) Network information propagation method based on length constraint and no repeated path
Liu et al. Finding densest lasting subgraphs in dynamic graphs: A stochastic approach
CN107784511A (en) A kind of customer loss Forecasting Methodology and device
US20080313359A1 (en) Relative typing waiting time before disambiguation aids
Wang et al. Password guessing using random forest
EP4322430A1 (en) Traffic detection method and apparatus, and device and readable storage medium
CN113518086B (en) Network attack prediction method, device and storage medium
CN113824797B (en) Self-adaptive synchronization method and device for teaching resources
Qin et al. Network traffic classification based on SD sampling and hierarchical ensemble learning
CN114065187A (en) Abnormal login detection method and device, computing equipment and storage medium
US9330063B2 (en) Generating a sparsifier using graph spanners
Tran et al. A Survey on Password Guessing
TWI689839B (en) Method for detecting suspicious degree of url link
JP2020119201A (en) Determination device, determination method and determination program
Zhang et al. Gaussian process learning for cyber‐attack early warning

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant