CN105228185A - A kind of method for Fuzzy Redundancy node identities in identification communication network - Google Patents
A kind of method for Fuzzy Redundancy node identities in identification communication network Download PDFInfo
- Publication number
- CN105228185A CN105228185A CN201510641602.XA CN201510641602A CN105228185A CN 105228185 A CN105228185 A CN 105228185A CN 201510641602 A CN201510641602 A CN 201510641602A CN 105228185 A CN105228185 A CN 105228185A
- Authority
- CN
- China
- Prior art keywords
- node
- fuzzy
- placeholder
- communication network
- matrix
- 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.)
- Granted
Links
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W24/00—Supervisory, monitoring or testing arrangements
- H04W24/04—Arrangements for maintaining operational condition
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W84/00—Network topologies
Landscapes
- Engineering & Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Data Exchanges In Wide-Area Networks (AREA)
Abstract
The invention belongs to the structural research field of complex network, be specifically related to a kind of method for Fuzzy Redundancy node identities in identification communication network.It is as follows that the method comprising the steps of, obtaining communication network data, communication network data is first distinguished known determining section and fuzzy uncertain part, for the transmit leg occurred in the communication information got, recipient and possibility be vacancy or indefinite transmit leg because of information dropout, recipient represents with a placeholder respectively, build a preliminary annexation figure, then according to the attribute of these placeholders and the topological characteristic in annexation figure, spectral clustering is adopted to carry out cluster to these placeholders, determine that in fact which placeholder represents same node, cluster is merged into a communication node to the placeholder in same group, thus reach and identify and remove fuzzy, the target of redundant node, realize the reconstruct of communication network.
Description
Technical field
The invention belongs to the structural research field of complex network, be specifically related to a kind of method for Fuzzy Redundancy node identities in identification communication network, be applicable to the reduction of network topology structure in communication network, Denoising Problems.
Background technology
Reconstruct for communication network topology structure is by imperfectly even analyzing containing noisy communication data of getting, thus restores as far as possible close to real network configuration.Communication network reconstruction comprises three aspects: the reduction of disappearance link; The reduction of disappearance, hiding communication node; And identity ambiguous, the distinguishing and removing of redundant node.Herein for the 3rd aspect, namely in communication network, the identification problem of fuzzy node is studied.
In the epoch of information explosion, particularly under complex electromagnetic environment, because the utilization of the means such as interference, camouflage, disappearance, the omission, even mistake of information is there is unavoidably in the communication information that we get, also have a large amount of redundancy noise, the communication information form that such as we get is generally: [call duration time, transmit leg, recipient, data package size], but because factors such as interference, noises, situations such as packet packet loss may being caused, get lines crossed, transmit leg recipient is indefinite or wrong.Section communication data may lose or damage, and the data source of the communication data that recipient is obtained may exist mistake, and many communications originally sent from same point may be mistaken as and send from two even multiple different nodes.This makes acceptance point accurately cannot judge to expect the state etc. of communication source.If have multiple node to there is this problem in a communication network, then can produce a large amount of lituras in a network, therefore the communication of whole network can be affected.The source node (node that fuzzy node should represent) differentiating these fuzzy nodes if can not divide, the communication efficiency of network and accuracy rate all can be affected.In order to maintain the stability of communication network, must distinguish fuzzy node, the association situation between communication network interior joint is returned to perfect condition as far as possible.
The domestic and international reconstruction about complex network topologies has carried out some researchs at present, mainly concentrates on disappearance link prediction aspect, and the relevant research about nodes information is just at the early-stage.Existing a small amount of research about communication network interior joint information only relates to the prediction of disappearance node, mainly contains two kinds of thinkings:
Simultaneously the first predict disappearance node and Lian Bian from the angle of network reconfiguration.Expect maximum framework and Kronecker graph model as [1] such as Leskovec combines, propose KronEM algorithm, first carry out estimation model parameter with the network observed, then model prediction lack part is used, again by the network configuration estimation model parameter of prediction, iteration like this, until parameter convergence.
The second is directly prediction disappearance node, and the people such as Wen-XuWang [2] utilize compressed sensing to carry out network reconfiguration in based on game theoretic network, thinks that reconstruct occurs that abnormal point is connected with certain implicit point, thus prediction missing point position in a network.Author defines a kind of game mechanism first in a network, and individuality has two kinds of selections, cooperation or betrayal (S={cooperation, Defense}), adopts the policy update that the simulation of Fermi rule is individual.Compressed sensing is be used in signal recuperation at first, restores complete signal by the partial data of the sparse signal got.For based on game theoretic evolved network, through too much taking turns game, because link probability matrix is a sparse matrix, meet compressed sensing definition, the relation of the strategy of often taking turns according to individuality and the income got and chain matrice, just change into the form of similar Y=Φ X, Y represents measured value, Φ represents calculation matrix, and X represents actual signal.In the link probability matrix finally obtained, if probability just thinks have limit to be connected close to 1, just think do not have limit to be connected close to 0.The threshold value of author's setting is 0.1.For taking turns game more, the link probability matrix that each compressed sensing reconstructs in theory should be the same, because the game mechanism in literary composition does not change network linking situation, if but the connection of certain several node to be taken turns and different and do not conform to the topological characteristic of network own at each, just can judge that these abnormity point are connected with missing point, thus determine missing point position in a network.
The people such as J.P.Bagrow [3] are from the dynamic behavior on network, for fixing network configuration, information flow situation on analog simulation network, then by the information flow speed determination information velocity of liquid assets of prediction algorithm prediction through each node, judge to occur the neighbours of abnormal point as missing point, thus determine missing point position in a network.
Although prior art can carry out preliminary judgement to lacking in communication network node, but the algorithm computation complexity of Leskovec is high especially, inapplicable in the real network, the method of compressed sensing is only applicable to the known network of mechanism of Evolution, algorithm more satisfactoryization of the people such as J.P.Bagrow, can only test on emulated data collection.Above-mentioned three algorithms are all that disappearance possible in initial analysis network implies node, and do not analyze redundant node possible in network, fuzzy node and identify.
List of references in literary composition:
[1]M.Kim,J.Leskovec.TheNetworkCompletionProblem:InferringMissingNodesandEdgesinNetworks.SIAMInternationalConferenceonDataMining(SDM)2011.
[2]Wen-XuWang,Ying-ChengLai,CelsoGrebogi,andJiepingYe.NetworkReconstructionBasedonEvolutionary-GameDataviaCompressiveSensing.PHYSICALREVIEWX1,021021(2011).
[3]J.P.Bagrow,S.Desu,M.R.Frank,N.Manukyan,L.Mitchell,A.Reagan,E.E.Bloedorn,L.B.Booker,L.K.Branting,M.J.Smith,B.F.Tivnan,C.M.Danforth,P.S.Dodds,andJ.C.Bongard,Shadownetworks:Discoveringhiddennodeswithmodelsofinformationflow.InProceedingsofCoRR.2013.
Summary of the invention
For in communication network because the factor such as interference, noise may occur packet packet loss, get lines crossed, the fuzzy node that situation causes such as transmit leg recipient is indefinite or wrong, redundant node identify, the invention enables the communication network topology reconstructed close to its real structure.Concrete technical scheme is as follows:
For a method for Fuzzy Redundancy node identities in identification communication network, comprise the following steps:
(1) obtaining communication network data, first distinguishes known determining section G by communication network data
k=<V
k, E
k> and fuzzy uncertain part, represent each uncertain node placeholder; Build a preliminary annexation figure G
a=<V
a, E
a>, wherein, <V
a, E
a>, <V
k, E
k> is connected respectively graph of a relation G
a, G
kmiddle comprised node set and Lian Bian set, calculate placeholder number | V
p|=| V
a|-| V
k|; V
krepresent known and determine node, V
prepresent placeholder, i.e. uncertain fuzzy node, V
arepresent whole node;
(2) adopt Gauss's distance calculating method, calculate preliminary annexation figure G
ain all nodes between incidence matrices
wherein | V
a| be node set V
ain element number;
represent real number field,
represent in real number field | V
a| row, | V
a| the set of column matrix.
(3) define
for diagonal matrix, wherein, diagonal element D
iifor the i-th row element sum of incidence matrices C, i=1 .., | V
a|, by
set up corresponding matrix
wherein
represent that each element of diagonal matrix D is made even the inverse of root;
(4) suppose that the source node number that fuzzy node should represent is known, be designated as h, find the maximal eigenvector of h L, composition matrix
wherein h the characteristic vector of L is respectively the row of Q;
(5) normalized matrix Q makes every behavior unit length, is designated as matrix Q ';
(6) remove the known row determining node of the middle correspondence of matrix Q ', retain the row of corresponding placeholder in matrix, obtain matrix
now, matrix Q " in the corresponding placeholder of every a line;
(7) Q " in row adopt k-mediods clustering to become h class;
(8) placeholder in same class is merged into a node, the node be merged into neighbours original with the placeholder in such are connected.
Further, Gauss's distance calculating method incidence matrices is adopted in described step (2)
process as follows:
Definition d
ifor preliminary annexation figure G
ainterior joint i to the vector of the shortest path length of other all nodes, then
Parameter σ is the standard deviation of Gauss's range formula, || || represent and ask vector field homoemorphism computing, e represents math constant, i.e. the truth of a matter of natural logrithm, C
ijthe element of the i-th row jth column position corresponding in representing matrix C.Because the larger sensitivity of σ lower (it is less on Gauss's distance impact between node that network increases and decreases a limit), σ more sluggishness is higher, and when value is between 3-5, discrimination better (in embodiment, the general value of σ is 4).
Adopt the beneficial effect that the present invention obtains: the present invention is used for data in communication network and lacks the imperfect inaccurate problem of network topology structure reconstruct caused, fuzzy node in communication network is distinguished and merged, to make the communication network topology reconstructed close to its real structure as far as possible, at present both at home and abroad also not for the research that Fuzzy Redundancy node problems in identification communication network is relevant, test the present invention can reach higher accuracy rate (more than 75%) for the identification of Fuzzy Redundancy node by experiment.By the present invention for the identification of Fuzzy Redundancy node and merging, the accuracy of communication network topology structural remodeling can being improved, providing authentic data to support for carrying out Analytic Network Process, communication process analysis, Business Process Analysis etc. further on a communication network.
Accompanying drawing explanation
Fig. 1 is that correct diagram of communications networks contrasts schematic diagram with containing Fuzzy Redundancy nodal communication network figure;
Fig. 2 is Clustering Effect schematic diagram;
Fig. 3 is the inventive method flow chart;
Fig. 4 is artificial network original graph;
Fig. 5 is network diagram after emulated data collection splits;
Fig. 6 is open Network data set Football primitive network figure.
Embodiment
Below, the invention will be further described with specific embodiment by reference to the accompanying drawings.
The present invention for the basic ideas of Fuzzy Redundancy node recognition problem in communication network is, outside the node that a removing part is determined, for the transmit leg occurred in the communication information got, recipient and possibility be vacancy or indefinite transmit leg because of information dropout, recipient represents with a placeholder respectively, build a preliminary annexation figure, then according to the attribute of these placeholders and the topological characteristic in annexation figure, spectral clustering is adopted to carry out cluster to these placeholders, determine that in fact which placeholder represents same node, cluster is merged into a communication node to the placeholder in same group, thus reach and identify and remove fuzzy, the target of redundant node, realize the reconstruct of communication network.
In the process that the structure that the communication data by getting is original to communication network is reduced, because noise, the factors such as interference, cause the imperfect even mistake of communication data, unavoidable in the network constructed exist some uncertain nodes, multiple redundant node may be easily mistaken for into, as shown in Figure 1, figure (a) represents correct diagram of communications networks, figure (b) represents in network containing Fuzzy Redundancy node schematic diagram, No. 6 in figure, No. 7 nodes are Fuzzy Redundancy node, by the algorithm in the present invention, cluster is carried out to the uncertain node of these redundancies, identify its real identity, as shown in Figure 2, thus restore as far as possible close to real communication network topology structure.
First the gatherer process of data is described below.Adopt emulated data collection and true social Network data set mutual authentication algorithm validity.The thinking of checking concentrates Stochastic choice part node to be split into multiple node (that is: redundant node) from emulated data and public data, if the node recognition after fractionation can be gone out its original corresponding node by algorithm, then prove that algorithm is effective.For the subgraph obtained, from whole node, a random selecting ξ node is as experimental point set V
m, V
mbe the source node that fuzzy node should represent, then this ξ point split into several respectively, form fuzzy set of node, with placeholder v ' ∈ V
preplace, V
prepresent placeholder collection, this just constitutes preliminary annexation figure G
a.Several its real representations of such fractionation node out source node, split all representing with placeholder out, whole data set processing procedure is exactly dataset construction process, constructs a training set for embodiment.The object of algorithm is that the placeholder (i.e. Fuzzy Redundancy node) split out is gone back to cluster, finds out which fuzzy node and is actually the same source node of expression.
Generate BA scales-free network with Matlab emulation, create-rule is BA (m0, m, N, pp), wherein m0 is the network node number before increasing, and m is limit number newly-generated when at every turn introducing new node, N is the network size after increasing, pp is initial network situation, and the value of pp has 1,2,3 three kinds, wherein 1 represents it is all isolated; 2 represent formation complete graph; 3 represent random connects some limits.Arrange different parameters respectively and generate two scales-free network, as shown in Figure 4, figure (a) is the corresponding parameter BA of network 1 (4,3,100,2); Figure (b) is the corresponding parameter BA of network 2 (10,6,100,3).
In experiment, in two networks, random selecting three points are as experimental point respectively, and what choose in network 1 is 9,19, No. 57 three nodes, are 69,87, No. 93 three nodes in network 2.Suppose that the communication information transmission carried out on that network has noise, there is information dropout in the communication that these nodes carry out, thus cause these nodes not occur in the network architecture, multiple different identity is marked as when carrying out Data Collection, node uncertain in network identifies by algorithm that the present invention proposes exactly, restores real network node.
In these two networks, find three nearest-neighbors of testing node respectively, then experiment node is split respectively, shown in corresponding relation following table.
Node corresponding relation is tested in table 1 emulated data
As shown in Figure 5, the node after fractionation is connected with the neighbours of former experiment node network diagram after fractionation respectively, the company limit of the former experiment node half number of each new reservation, preliminary annexation figure G
a.
Public data integrates as Football network, and this data set record participates in the signing situation of 22 football team members between 35 countries of world cup for 1998, and the company limit in network represents that certain member outputs to another country from a country.This data set is regarded as and haves no right Undirected networks, comprise 35 nodes, 118 limits.Initial network connection layout as shown in Figure 6.Arbitrarily choose No. 3, No. 14, No. 16 three nodes as experiment node, split by these three nodes respectively, each node splits into four parts, split posterior nodal point and origin node corresponding relation as shown in table 2.
Table 2Football data set experiment node corresponding relation
As shown in Figure 3, be flow chart of the present invention; Specific embodiment one, for Football Network data set (in experiment, σ gets 4):
(1) for the experimental data collection constructed, known determining section G is first distinguished
k=<V
k, E
k> and fuzzy uncertain part, in its Central Plains network, in 35 nodes, except 3,14, No. 16 nodes, (these 3 node compositions gather V
m) beyond 32 nodes form known determining section V
k, 36 to No. 47 that split out have 12 nodes altogether is Fuzzy Redundancy node V
p, each Fuzzy Redundancy node, as 1 placeholder, builds preliminary annexation figure G
a=<V
a, E
a>, <V
a, E
a> is respectively annexation figure G
athe node set comprised and Lian Bian set, V
aknownly determine node V
kwith Fuzzy Redundancy node V
punion, E
arepresent V
aannexation between interior joint;
(2) adopt Gauss's distance calculating method, calculate preliminary annexation figure G
ain all nodes between incidence matrices
wherein 44 is set V
ain element number, namely G
ain the nodes that comprises;
(3) diagonal element D
iithe i-th row element sum that (i=1 .., 44) are incidence matrices C, by
set up corresponding matrix
(4) the fuzzy node source node number h=3 that should represent, finds the maximal eigenvector of 3 L (looking for vertical characteristic vector to multiple eigenvalue), composition matrix
wherein 3 characteristic vectors of L are respectively the row of Q;
(5) standardization Q matrix makes every behavior unit length, is designated as Q ';
(6) (algorithm is according to node serial number a line by each node homography, corresponding i-th row of such as node i, is corresponding front 32 row in the present embodiment to remove in Q ' matrix the corresponding known row determining node; Retaining the row of corresponding placeholder in matrix, is last 12 row of matrix in the present embodiment), obtain matrix
now, Q " the corresponding placeholder of every a line in matrix;
(7) Q " in 12 row adopt k-mediods method (a kind of existing sample clustering method) be clustered into 3 classes, namely carry out cluster to No. 36 to No. 47 these 12 placeholders, cluster result is:
Table 3 placeholder cluster result
36 38 39 43 47 | 37 40 41 42 | 44 45 46 |
(8) placeholder in same class is merged into a node, the neighbours original with the placeholder in such are connected.The result that cluster obtains, namely the corresponding relation of the source node that obtains of cluster and Fuzzy Redundancy node is as following table:
Node corresponding relation tested by table 4
Node serial number | 36 38 39 43 47 | 37 40 41 42 | 44 45 46 |
Obtain corresponding source node numbering | 3 | 14 | 16 |
The accuracy of cluster is 9/12=0.75, illustrates that this algorithm effectively can identify the true identity of the uncertain node of redundancy in network as can be seen from the results.
Specific embodiment two, tests according to step emulated data of the present invention.Adopt the algorithm in the present invention to analyze the uncertain node emulated in two networks obtaining, cluster result is as following table:
Table 5 emulated data predicts the outcome
Node serial number | 101 | 102 | 103 | 104 | 105 | 106 | 107 | 108 | 109 | 110 | Accuracy |
Network 1 | 9 | 57 | 19 | 9 | 57 | 57 | 57 | 57 | 19 | 19 | 0.8 |
Network 2 | 69 | 69 | 69 | 69 | 93 | 93 | 87 | 87 | 87 | 87 | 1 |
The grouping that after secondary series, the 3rd row are respectively cluster in table, fuzzy redundant node is corresponding, experiment node namely in corresponding former network, can find out, for network 1, algorithm predicts accuracy reaches 0.8, the prediction accuracy of network 2 reaches 1.0, illustrates that this algorithm effectively can identify the true identity of the uncertain node of redundancy in network.
More than to invention has been exemplary description; obvious realization of the present invention is not subject to the restrictions described above; as long as have employed the various improvement that technical solution of the present invention is carried out; or design of the present invention and technical scheme directly applied other occasion, all in protection scope of the present invention without to improve.
Claims (2)
1., for a method for Fuzzy Redundancy node identities in identification communication network, it is characterized in that, comprise the following steps:
(1) obtaining communication network data, first distinguishes known determining section G by communication network data
k=<V
k, E
k> and fuzzy uncertain part, represent each uncertain node placeholder; Build a preliminary annexation figure G
a=<V
a, E
a>, wherein, <V
a, E
a>, <V
k, E
k> is connected respectively graph of a relation G
a, G
kmiddle comprised node set and Lian Bian set, calculate placeholder number | V
p|=| V
a|-| V
k|; V
krepresent known and determine node, V
prepresent placeholder, i.e. uncertain fuzzy node, V
arepresent whole node;
(2) adopt Gauss's distance calculating method, calculate preliminary annexation figure G
ain all nodes between incidence matrices
wherein | V
a| be node set V
ain element number,
represent real number field;
(3) define
for diagonal matrix, wherein, diagonal element D
iifor the i-th row element sum of incidence matrices C, i=1 .., | V
a|, by
set up corresponding matrix
wherein
represent that each element of diagonal matrix D is made even the inverse of root;
(4) suppose that the source node number that fuzzy node should represent is known, be designated as h, find the maximal eigenvector of h L, composition matrix
wherein h the characteristic vector of L is respectively the row of Q;
(5) normalized matrix Q makes every behavior unit length, is designated as Q ';
(6) remove the known row determining node of the middle correspondence of matrix Q ', retain the row of corresponding placeholder in matrix, obtain matrix
now, matrix Q " in the corresponding placeholder of every a line;
(7) Q " in row adopt k-mediods clustering to become
hindividual class;
(8) placeholder in same class is merged into a node, the neighbours original with the placeholder in such are connected.
2. a kind of method for Fuzzy Redundancy node identities in identification communication network as claimed in claim 1, is characterized in that, adopts Gauss's distance calculating method incidence matrices in described step (2)
process as follows:
Definition d
ifor preliminary annexation figure G
ainterior joint i to the vector of the shortest path length of other all nodes, then
Parameter σ is the standard deviation of Gauss's range formula, || || represent and ask vector field homoemorphism computing.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201510641602.XA CN105228185B (en) | 2015-09-30 | 2015-09-30 | Method used for identifying identity of fuzzy redundant node in communication network |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201510641602.XA CN105228185B (en) | 2015-09-30 | 2015-09-30 | Method used for identifying identity of fuzzy redundant node in communication network |
Publications (2)
Publication Number | Publication Date |
---|---|
CN105228185A true CN105228185A (en) | 2016-01-06 |
CN105228185B CN105228185B (en) | 2017-03-22 |
Family
ID=54996797
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201510641602.XA Active CN105228185B (en) | 2015-09-30 | 2015-09-30 | Method used for identifying identity of fuzzy redundant node in communication network |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN105228185B (en) |
Cited By (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN106021487A (en) * | 2016-05-19 | 2016-10-12 | 浙江工业大学 | Internet of Things semantic event detection method based on fuzzy theory |
CN109190204A (en) * | 2018-08-15 | 2019-01-11 | 贵州大学 | A kind of complex mechanical product module partition method based on complex network |
CN112104515A (en) * | 2020-11-19 | 2020-12-18 | 中国人民解放军国防科技大学 | Network reconstruction method and device, computer equipment and storage medium |
CN112256941A (en) * | 2020-10-20 | 2021-01-22 | 北京工商大学 | Social network reconstruction method based on local information |
CN112994991A (en) * | 2021-05-20 | 2021-06-18 | 中南大学 | Redundant node discrimination method, device and equipment and readable storage medium |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101556651A (en) * | 2009-04-15 | 2009-10-14 | 北京航空航天大学 | Multi-source data fusion method in clustering wireless sensor network |
CN101795501A (en) * | 2009-10-30 | 2010-08-04 | 中国科学院上海微系统与信息技术研究所 | Physical signal collaborative compression sensing system and method for sensor network |
CN102170648A (en) * | 2011-01-28 | 2011-08-31 | 北京浩阳华夏科技有限公司 | Passive diagnosis method of wireless sensor network |
CN102662988A (en) * | 2012-03-14 | 2012-09-12 | 中山大学 | Method for filtering redundant data of RFID middleware |
CN104579790A (en) * | 2015-01-26 | 2015-04-29 | 中国人民解放军国防科学技术大学 | Method for determining quantity of link prediction restitution edges |
-
2015
- 2015-09-30 CN CN201510641602.XA patent/CN105228185B/en active Active
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101556651A (en) * | 2009-04-15 | 2009-10-14 | 北京航空航天大学 | Multi-source data fusion method in clustering wireless sensor network |
CN101795501A (en) * | 2009-10-30 | 2010-08-04 | 中国科学院上海微系统与信息技术研究所 | Physical signal collaborative compression sensing system and method for sensor network |
CN102170648A (en) * | 2011-01-28 | 2011-08-31 | 北京浩阳华夏科技有限公司 | Passive diagnosis method of wireless sensor network |
CN102662988A (en) * | 2012-03-14 | 2012-09-12 | 中山大学 | Method for filtering redundant data of RFID middleware |
CN104579790A (en) * | 2015-01-26 | 2015-04-29 | 中国人民解放军国防科学技术大学 | Method for determining quantity of link prediction restitution edges |
Cited By (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN106021487A (en) * | 2016-05-19 | 2016-10-12 | 浙江工业大学 | Internet of Things semantic event detection method based on fuzzy theory |
CN106021487B (en) * | 2016-05-19 | 2019-05-28 | 浙江工业大学 | A kind of Internet of Things semantic events detection method based on fuzzy theory |
CN109190204A (en) * | 2018-08-15 | 2019-01-11 | 贵州大学 | A kind of complex mechanical product module partition method based on complex network |
CN112256941A (en) * | 2020-10-20 | 2021-01-22 | 北京工商大学 | Social network reconstruction method based on local information |
CN112256941B (en) * | 2020-10-20 | 2023-10-13 | 北京工商大学 | Social network reconstruction method based on local information |
CN112104515A (en) * | 2020-11-19 | 2020-12-18 | 中国人民解放军国防科技大学 | Network reconstruction method and device, computer equipment and storage medium |
CN112994991A (en) * | 2021-05-20 | 2021-06-18 | 中南大学 | Redundant node discrimination method, device and equipment and readable storage medium |
CN112994991B (en) * | 2021-05-20 | 2021-07-16 | 中南大学 | Redundant node discrimination method, device and equipment and readable storage medium |
Also Published As
Publication number | Publication date |
---|---|
CN105228185B (en) | 2017-03-22 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Zhang et al. | Identification of core-periphery structure in networks | |
CN105228185A (en) | A kind of method for Fuzzy Redundancy node identities in identification communication network | |
CN109948000B (en) | Abnormal target detection method, device, equipment and storage medium for heterogeneous network | |
CN110147911B (en) | Social influence prediction model and prediction method based on content perception | |
CN102495919B (en) | Extraction method for influence factors of carbon exchange of ecosystem and system | |
CN112615888B (en) | Threat assessment method and device for network attack behavior | |
CN106470122A (en) | A kind of network failure locating method and device | |
CN102879677A (en) | Intelligent fault diagnosis method based on rough Bayesian network classifier | |
Xiang et al. | Acquisition of causal models for local distributions in Bayesian networks | |
CN106960251A (en) | A kind of Undirected networks based on node similitude connect side right value Forecasting Methodology | |
CN106327345A (en) | Social group discovering method based on multi-network modularity | |
CN103559205A (en) | Parallel feature selection method based on MapReduce | |
CN106997373A (en) | A kind of link prediction method based on depth confidence network | |
CN111537884A (en) | Method and device for acquiring service life data of power battery, computer equipment and medium | |
Xiao et al. | Network security situation prediction method based on MEA-BP | |
CN105068928A (en) | Complex network theory based software test use-case generating method | |
CN106912040B (en) | Ad Hoc network key node identification method fusing deletion method | |
CN115051929A (en) | Network fault prediction method and device based on self-supervision target perception neural network | |
Shi et al. | Network reliability analysis through survival signature and machine learning techniques | |
CN112910865B (en) | Inference attack stage maximum likelihood estimation method and system based on factor graph | |
Enikeeva et al. | Change-point detection in dynamic networks with missing links | |
Prasetyowati et al. | Dataset feasibility analysis method based on enhanced adaptive LMS method with Min-max normalization and fuzzy intuitive sets | |
Tao et al. | Metric ranking of invariant networks with belief propagation | |
CN109492677A (en) | Time-varying network link prediction method based on bayesian theory | |
CN112766320B (en) | Classification model training method and computer equipment |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
C14 | Grant of patent or utility model | ||
GR01 | Patent grant |