CN104539731B - A kind of trusted service method of diffusion and device based on community network node optimizing - Google Patents

A kind of trusted service method of diffusion and device based on community network node optimizing Download PDF

Info

Publication number
CN104539731B
CN104539731B CN201510023305.9A CN201510023305A CN104539731B CN 104539731 B CN104539731 B CN 104539731B CN 201510023305 A CN201510023305 A CN 201510023305A CN 104539731 B CN104539731 B CN 104539731B
Authority
CN
China
Prior art keywords
node
service
diffusion
degree
trusted
Prior art date
Application number
CN201510023305.9A
Other languages
Chinese (zh)
Other versions
CN104539731A (en
Inventor
张佩云
Original Assignee
安徽师范大学
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 安徽师范大学 filed Critical 安徽师范大学
Priority to CN201510023305.9A priority Critical patent/CN104539731B/en
Publication of CN104539731A publication Critical patent/CN104539731A/en
Application granted granted Critical
Publication of CN104539731B publication Critical patent/CN104539731B/en

Links

Classifications

    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04LTRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L67/00Network-specific arrangements or communication protocols supporting networked applications
    • H04L67/32Network-specific arrangements or communication protocols supporting networked applications for scheduling or organising the servicing of application requests, e.g. requests for application data transmissions involving the analysis and optimisation of the required network resources

Abstract

The invention discloses a kind of trusted service method of diffusion based on community network node optimizing and device, methods described to include:Success rate is spread according to community network node serve, it is determined that the degree of belief of each community network node;According to the degree of belief of each community network node, it is determined that whether each community network node is trusted node;For identified trusted node, diffusion number of success and success rate, the associated volume of trusted node and neighbor node, neighbor node quantity are serviced according to trusted node, determine that trusted node services diffusing capacity;Diffusing capacity is serviced according to trusted node, determines whether trusted node is advantage node, so that advantage node carries out trusted service diffusion determined by.The present invention can improve trusted service diffusion velocity.

Description

A kind of trusted service method of diffusion and device based on community network node optimizing

Technical field

It is more particularly to a kind of based on the credible of community network node optimizing the present invention relates to social network service diffusion technique Service method of diffusion and device.

Background technology

As service-oriented software architecture, software service (Software-as-a-Service SaaS), cloud The development of new software architecture thought and the Operational ideas such as calculating, service provide as computing resource important in internet and software Production, it is increasingly paid attention to.

Service be unable to do without issue, finds and integrated, conventional method using web service architecture centralization Mechanism, such as universal description, discovery and integrated service (Universal Description Discovery and Integration, UDDI), the mechanism, which is faced with the data volume for needing to manage and moment, needs request of data to be processed constantly to increase The problems such as more and registrar is also highly prone to the attack of some malicious nodes, so as to cause to influence the bottleneck of service performance, And there is scalability and poor reliability, be difficult to find the interactive relation between service the problems such as, prior art also indicates that concentration How formula service registry stock improves the efficiency and validity of trusted service diffusion in Problem of Failure, to promote the quick of service Shared and efficient utilize is current important subject.With the fast development of the community networks such as Facebook, Twitter, in advance Show that Social Media turns into the focus and trend of current internet development, it is just gradually affecting the life of the network user, study And work.The optimization method of service discovery in community network is studied, and trusted service is carried out by the trusting relationship of community network Diffusion is just turning into the quick effective way for finding trusted service in community network.

Due to traditional service method of diffusion often run into when handling large scale community network data computational efficiency and The bottlenecks such as algorithm performance, and the structure of community network has scalability, therefore, can by finding optimal community network node Trusted service is promoted to share.

The content of the invention

It is an object of the invention to provide a kind of trusted service method of diffusion and device based on community network node optimizing, Can solve the problems, such as the fast and effective diffusion of trusted service better by optimal community network node is found.

A kind of according to an aspect of the invention, there is provided trusted service diffused sheet based on community network node optimizing Method, including:

Success rate is spread according to community network node serve, it is determined that the degree of belief of each community network node;

According to the degree of belief of each community network node, it is determined that whether each community network node is trusted node;

For identified trusted node, according to trusted node service diffusion number of success and success rate, trusted node with Associated volume, the neighbor node quantity of neighbor node, determine the service diffusing capacity of trusted node;

Diffusing capacity is serviced according to trusted node, determines whether trusted node is advantage node, so as to excellent determined by utilizing Gesture node carries out trusted service diffusion.

Preferably, some community network node v is determined by following stepsjDegree of belief:

According to neighbor node viTo vjThe service diffusion success rate of diffusion service, determines viTo vjDegree of belief;

Utilize identified viTo vjDegree of belief and viWeight, determine vjDegree of belief.

Preferably, v is determined by following stepsjWhether it is trusted node:

By vjDegree of belief compared with default degree of belief threshold value;

Work as vjDegree of belief when being more than default degree of belief threshold value, determine vjFor trusted node, v is otherwise determinedjFor insincere section Point.

Preferably, when it is determined that vjFor trusted node when, determine v by following stepsjService diffusing capacity:

According to vjTo viThe service diffusion number of success of diffusion service, determines vjTo viSpread the service diffusion weightings of service;

According to vjWith viAssociated volume, determine vjThe degree of association;

Utilize vjTo viThe service diffusion success rate of diffusion service, vjThe degree of association, determine vjService diffusion average speed;

Using the service diffusion weightings, service diffusion average speed, neighbor node quantity, v is determinedjService diffusion Amount.

Preferably, v is determined by following stepsjWhether it is advantage node:

By vjDiffusing capacity is serviced compared with default diffusing capacity threshold value;

Work as vjWhen servicing diffusing capacity more than default diffusing capacity threshold value, v is determinedjFor advantage node, v is otherwise determinedjFor common section Point.

According to another aspect of the present invention, there is provided a kind of trusted service diffusion dress based on community network node optimizing Put, including:

Degree of belief determining module, for spreading success rate according to community network node serve, it is determined that each community network section The degree of belief of point;

Trusted node determining module, for the degree of belief according to each community network node, it is determined that each community network section Whether point is trusted node;

Diffusing capacity determining module, for identified trusted node, according to trusted node service diffusion number of success and into Power, the associated volume of trusted node and neighbor node, neighbor node quantity, determine the service diffusing capacity of trusted node;

Advantage node determining module, for servicing diffusing capacity according to trusted node, determine whether trusted node is advantage section Point, so that advantage node carries out trusted service diffusion determined by.

Preferably for some community network node vj, the degree of belief determining module is according to neighbor node viTo vjExpand The service diffusion success rate of service is dissipated, determines viTo vjDegree of belief, and utilize determined by viTo vjDegree of belief and viPower Weight, determines vjDegree of belief.

Preferably, the trusted node determining module is by vjDegree of belief compared with default degree of belief threshold value, work as vj's When degree of belief is more than default degree of belief threshold value, v is determinedjFor trusted node, v is otherwise determinedjFor insincere node.

Preferably, when it is determined that vjFor trusted node when, it is described service diffusing capacity determining module according to vjTo viDiffusion service Service diffusion number of success, determines vjTo viSpread the service diffusion weightings of service;According to vjWith viAssociated volume, determine vjPass Connection degree;Utilize vjTo viThe service diffusion success rate of diffusion service, vjThe degree of association, determine vjService diffusion average speed, and profit With the service diffusion weightings, service diffusion average speed, neighbor node quantity, v is determinedjService diffusing capacity.

Preferably, the advantage node determining module is by vjDiffusing capacity is serviced compared with default diffusing capacity threshold value, works as vj When servicing diffusing capacity more than default diffusing capacity threshold value, v is determinedjFor advantage node, v is otherwise determinedjFor ordinary node.

Compared with prior art, the beneficial effects of the present invention are:

Optimal service diffusion node (i.e. advantage node) is combined by the present invention with trusted service, on the one hand can improve service The accuracy rate and efficiency of diffusion, it on the other hand can improve the credibility of service diffusion.

Brief description of the drawings

Fig. 1 is the trusted service method of diffusion principle frame provided in an embodiment of the present invention based on community network node optimizing Figure;

Fig. 2 is the trusted service disperser structure chart provided in an embodiment of the present invention based on community network node optimizing;

Fig. 3 is advantage node, ordinary node and trusted service relation schematic diagram provided in an embodiment of the present invention;

Fig. 4 is the trusted service diffusion flow chart provided in an embodiment of the present invention based on community network node optimizing;

Fig. 5 is time performance analysis figure provided in an embodiment of the present invention;

Fig. 6 is trusted service diffusivity comparison schematic diagram provided in an embodiment of the present invention.

Embodiment

Below in conjunction with accompanying drawing to a preferred embodiment of the present invention will be described in detail, it will be appreciated that described below is excellent Select embodiment to be merely to illustrate and explain the present invention, be not intended to limit the present invention.

Fig. 1 is the trusted service method of diffusion principle frame provided in an embodiment of the present invention based on community network node optimizing Figure, as shown in figure 1, step includes:

Step S101:Success rate is spread according to community network node serve, it is determined that the degree of belief of each community network node.

With some community network node vjExemplified by, according to neighbor node viTo vjThe service diffusion success rate of diffusion service, Determine viTo vjDegree of belief, and utilize determined by viTo vjDegree of belief and viWeight, determine vjDegree of belief.

Wherein, vjThere is v1,v2,...,vNN number of neighbor node, i.e. vi, i=1,2 ..., N.

Wherein, neighbor node viTo vjThe service diffusion success rate of diffusion service passes through viTo vjService diffusion and obtain The service diffusion number of success and the ratio of total degree arrived.

Wherein, by the weight of each neighbor node and to vjDegree of belief product addition, then divided by each neighbours section The weight sum of point, obtains vjDegree of belief.

Step S102:According to the degree of belief of each community network node, it is determined that whether each community network node is credible Node.

By vjDegree of belief compared with default degree of belief threshold value, work as vjDegree of belief when being more than default degree of belief threshold value, Determine vjFor trusted node, v is otherwise determinedjFor insincere node.

Step S103:For identified trusted node, diffusion number of success and success rate are serviced according to trusted node, can Believe associated volume, the neighbor node quantity of node and neighbor node, determine the service diffusing capacity of trusted node.

First, according to vjTo viThe service diffusion number of success of diffusion service, determines vjTo viThe service diffusion of diffusion service Weight;According to vjWith viAssociated volume, determine vjThe degree of association, specifically, by vjWith the associated volume between N number of neighbor node The most relevance amount that (transaction relationship being present) divided by N number of neighbor node associate two-by-two, obtains vjTo viThe degree of association.Secondly, profit Use vjTo viThe service diffusion success rate of diffusion service, vjThe degree of association, determine vjService diffusion average speed, and utilize vjTo vi Spread service diffusion weightings, the v of servicejService diffusion average speed, neighbor node quantity, determine vjService diffusing capacity.

Step S104:Diffusing capacity is serviced according to trusted node, determines whether trusted node is advantage node, to utilize institute The advantage node of determination carries out trusted service diffusion.

By vjDiffusing capacity is serviced compared with default diffusing capacity threshold value, works as vjService diffusing capacity and be more than default diffusing capacity threshold During value, v is determinedjFor advantage node, v is otherwise determinedjFor ordinary node.

Fig. 2 is the trusted service disperser structure provided in an embodiment of the present invention based on community network node optimizing Figure, as shown in Fig. 2 including degree of belief determining module 10, trusted node determining module 20, diffusing capacity determining module 30 and advantage section Point determining module 40.

Degree of belief determining module 10 is used to spread success rate according to community network node serve, it is determined that each community network section The degree of belief of point.

Trusted node determining module 20 is used for the degree of belief according to each community network node, it is determined that each community network section Whether point is trusted node.

Diffusing capacity determining module 30 for identified trusted node, according to trusted node service diffusion number of success and into Power, the associated volume of trusted node and neighbor node, neighbor node quantity, determine the service diffusing capacity of trusted node.

Advantage node determining module 40 is used to service diffusing capacity according to trusted node, determines whether trusted node is advantage section Point, so that advantage node carries out trusted service diffusion determined by.

For some community network node vj, degree of belief determining module 10 is according to neighbor node viTo vjDiffusion service Service diffusion success rate, determines viTo vjDegree of belief, and utilize determined by viTo vjDegree of belief and viWeight, determine vj Degree of belief.Trusted node determining module 20 is by v determined by degree of belief determining module 10jDegree of belief and default degree of belief threshold Value is compared, and works as vjDegree of belief when being more than default degree of belief threshold value, determine vjFor trusted node, v is otherwise determinedjFor can not Believe node.If trusted node determining module 20 determines vjFor trusted node, then diffusing capacity determining module 30 is serviced according to vjTo vi The service diffusion number of success of diffusion service determines vjTo viThe service diffusion weightings of service are spread, according to vjWith viAssociated volume Determine vjThe degree of association, and utilize vjTo viThe service diffusion success rate of diffusion service, vjThe degree of association, determine vjService diffusion Average speed, finally using the service diffusion weightings, service diffusion average speed, neighbor node quantity, determine vjService Diffusing capacity.Advantage node determining module 40 will service v determined by diffusing capacity determining module 30jService diffusing capacity and default diffusion Amount threshold value is compared, and works as vjWhen servicing diffusing capacity more than default diffusing capacity threshold value, v is determinedjFor advantage node, v is otherwise determinedj For ordinary node.It is determined that vjAfter advantage node, you can fast and effeciently carry out trusted service expansion using the advantage node Dissipate.

Fig. 3 to Fig. 6 is combined below by way of correlation model and definition, the present invention is described in detail.

User node collection is represented with U in community network;Services set in community network is represented with S.U × S → R, R are one Serially ordered set, it is a utility function r (u, s), for calculating service s to node u credibility, for u ∈ U, in community network Trusted service diffusion can be with formalization representation such as formula (1):

By community network to the maximum trusted service that user u spreads be smax,u

1. service trust graph of a relation in community network

Service trust graph of a relation is represented with digraph G in community network, and G=(V, S, E, F), wherein vertex set V represent society Meeting network node collection, V={ v1,v2,v3,...,vn, n is integer.

Side collection E represents the transaction relationship between community network node, if community network node viAnd vjBetween exist one From viPoint to vjDirected edge (vi,vj) ∈ E, represent there is transaction between two nodes, use wijRepresent the weight of two nodes transaction, wij=< αijij>, wherein, αijWith βijCommunity network node v is represented respectivelyiTo node vjService diffusion and count obtained success Number and total degree.

S represents services set, S={ s1,s2,s3,...,sm, m is integer.

F represents there is transaction relationship and trust value between community network node and service, if community network node viWith service sk Between have one from viPoint to skDirected edge (vi,sk) ∈ F, represent community network node viWith servicing skBetween existed it is straight Service call transaction relationship is connect, uses fikRepresent community network node viTo servicing skTrust value.

This social network service trusting relationship has asymmetry.

2. community network node

Community network node is designated as v, v=(ID, NeighborList, Level, SList), wherein, ID represents social network The identifier of network node, NeighborList represent node v neighbours (Neighbor) set, NeighborList= { (NeighborID, NeighborDes) }, NeighborID represent the neighbor node identifier of the node, NeighborDes tables Show the description of neighbor node.Level is used for the grade for representing community network node, shown in its value condition such as formula (2).

The set of the used service of tune of SList expression community network nodes, SList=(SID, URI, Class, DES, Inv, Tot, Tag) }, SID represents the identifier of certain service, and URI represents the resource identifier of the service, and Class is the service institute The classification of category, DES are that the function of the service describes, and Inv represents the number that service is correctly called, and Tot represents to call service Total degree, Tag represent whether service allows to spread to its neighbor node, shown in its value condition such as formula (3).

3. the degree of belief of community network node

Community network node viTo node vjDirect degree of belief be designated as trusti[j], refer to based on community network node viWith Node vjTrading situation calculate viTo vjDegree of belief.viFor vjNeighbor node and be present in vjNeighbor list in. trustiThe value of [j] depends primarily on αijijAnd its result of dynamic regulation, wherein, αijWith βijCommunity network section is represented respectively Point viTo node vjService diffusion and count obtained number of success and total degree.

4. community network node vjDegree of belief

Community network node vjDegree of belief be designated as trust [vj], pass through vjAll neighbor node vi, utilize viTo vj's Direct degree of belief trusti[j] carries out aggregate operation and its dynamic feedback adjustment, calculates vjDegree of belief trust [vj].Wherein, Shown in aggregate operation formula such as formula (4):

In formula (4), N=| vj.neightlist |, N represents vjThe number of neighbor node, wiExpression represents node vi's Weight.

Community network node plays not there is strong and weak different, diverse difference in Service Source shared procedure Same action effect.To improve the shared performance of Service Source, the present invention community network node be divided into advantage Node subsets with Two subsets of ordinary node subset, advantage Node subsets are obtained by node optimizing, the shared of service is improved by advantage node Diffusivity and reliability.For relation between advantage node, ordinary node and trusted service as shown in figure 3, in Fig. 3, thick side is oval Ordinary node is represented, thin side ellipse representation advantage node, thin side is circular to represent service, → represent to service between community network node The direction of diffusion of information, the transaction weight between label w expression nodes thereon;- → represent community network node and the connection of service System, such as call relation, label f thereon represent trust weight of the community network node to the service.

The present invention is when carrying out community network node optimizing, both in view of the degree of belief of community network node, it is also contemplated that Service diffusion velocity and diffusing capacity, its reason are:(1) due to and not all community network node be all good will node, society Network often there is non-good will node, if its malice spreads a large amount of information on services, will cause blocking and the confusion of network; (2) an important factor for diffusion velocity and diffusing capacity are influence service diffusion effects is serviced, and due to the topological structure of community network And its continuous change of registration service information, aggregation extent between by community network node of service diffusion velocity and diffusing capacity and The influence of the node degree of association.

5. community network concentration class.

Community network concentration class is designated as η, if community network node number is a, the side number between a node is preferably at most a* (a-1) the side number, and actually between node is b, then community network concentration class η=b/ (a* (a-1)).

Community network concentration class is bigger, represents that the average degree of association between the node of the community network is stronger.

6. the degree of association of community network node.

Community network node viThe degree of association be designated as μiIf community network node viNeighbor node number be n, n node Between side number be preferably at most n* (n+1), and actually viSide number between n neighbor node is m, then node viThe degree of association For μi=m/ (n* (n+1)).

The node degree of association is a key character possessed by advantage node, and the bigger node of the node degree of association, representing should Node is connective stronger with other community network nodes.

Because the topological structure of community network is in continuous change, therefore, determine to save by giving a dilation angle R Relation between point, under dilation angle R, community network node viService diffusion average speed be defined as follows.

7. community network node viService average diffusion speed

Community network node viService average diffusion speed be designated as Ui R:Refer in given dilation angle Zone R domain, 0 ~T time section is interior, community network node viService average diffusion speed, can be measured by a statistical model, it is such as public Shown in formula (5).

In formula (5), μiRepresent community network node viThe degree of association, R represents dilation angle, and r is variable, 1≤r≤R, nrWhen expression dilation angle is r, the number of neighbor node in annular region of the radius between r and radius (r-1);Work as viFor source During node, αj rExpression radius be r- (r-1) region, node viSucceed to its neighbor node vjService the number of diffusion; βj rRepresent that radius is r to radius in the annular region of (r-1), node viTo its neighbor node vjService total time of diffusion Number.αj rj rRepresent viTo neighbor node vjThe success rate of information on services is spread,Represent viTo all neighbor nodes Spread the success rate of information on services.Represent in R radius regions and in 0~T time section internal segment Point viService diffusion average speed.The value is bigger, illustrates community network node viService diffusion average speed it is faster.Formula (5) in, as R=r=1, viFor source node, thenWhen 1<During r≤R, then in r=1 base On plinth, recursively by community network node viNeighbor node regard source node as, until r be equal to dilation angle R untill.

Formula (5) overcomes the locality characteristics of node, considers not only the service average diffusion speed of node itself, In view of service diffusion velocity of the node to the regional nodes where specified dilation angle, therefore, the average speed is not The local average speed of node, but the average speed in the diffusion zone of radius is specified, therefore, the average speed has preferable It is of overall importance.

8. community network node viService diffusing capacity

Community network node viService diffusing capacity be designated as ψ(T)(vi), refer to community network node viIn 0~T time section By viInformation on services be diffused into vjThe information on services amount size of neighbor node, as shown in formula (6):

In formula (6), N represents viNeighbor node number;Ui RAverage speed is spread for the service of community network node; wij TRepresent 0~T time section viTo vjService diffusing capacity weight, as shown in formula (7):

In formula (7),Represent t node viTo N number of neighbor node vjThe total degree of service is spread in success;Represent t community network node (vi,vj) between stateful transaction value, shown in value such as formula (8):

In formula (8), in initial time, by the source node v of information on servicesiTo community network node vjStateful transaction δi t [j] is arranged to 0.Node viInformation on services is diffused into vjNode, if spread successfully, δi t[j] is changed into 1 from 0, otherwise state Still it is 0.If from node viWith node vjBetween not have to spread between side or both an information on services unsuccessful, then δi tThe value of [j] will It is 0 always.If having ring between two nodes, need first to determine the stateful transaction, another node etc. of the node section Point determines whether just to change state after receiving information on services, so as to avoid that deadlock occurs.

Fig. 4 is the trusted service diffusion flow chart provided in an embodiment of the present invention based on community network node optimizing, is being taken , it is necessary to carry out node optimizing during business diffusion, that is, the big node of service diffusing capacity is found, the present embodiment is with some community network section Point viExemplified by, viWith n neighbor node vj, i=1,2 ..., n, as shown in figure 4, step includes:

Step S401:Calculate community network node viDegree of belief trust [vi]。

First, according to all neighbor node vjTo viThe service diffusion success rate of diffusion service, determines vjTo viDegree of belief trustj[i], specifically, trustjThe value of [i] depends primarily on αjijiAnd its result of dynamic regulation, wherein, αjiWith βji Community network node v is represented respectivelyjTo node viService diffusion and count obtained number of success and total degree.

Then, all neighbor node v are utilizedjTo viDegree of belief trustj[i] carries out aggregate operation, calculates viTrust Spend trust [vi], specifically,Wherein, wiExpression represents node viWeight.

Step S402:Judge whether community network node meets degree of belief threshold requirement, i.e., by viDegree of belief with it is default Degree of belief threshold value is compared, if viDegree of belief be more than default degree of belief threshold value, then perform step S403 successively to step S405, otherwise perform step S409.

Step S403:Work as viDegree of belief when being more than default degree of belief threshold value, determine viFor trusted node.

Step S404:Calculate node viService average diffusion speed Ui RDiffusing capacity weight w is serviced between nodeij T

For servicing diffusing capacity weight w between nodeij T, according to 0~T time section, viTo vjThe service diffusion of diffusion service Number of success, determine viTo vjSpread the service diffusing capacity weight w of serviceij T, specifically,Wherein,Represent t node viTo N number of neighbours Node vjThe total degree of service, δ are spread in successi t[j] represents t community network node (vi,vj) between stateful transaction value,

For node viService average diffusion speed Ui R, first, calculate community network node viDegree of association μi=m/ (n* (n+1)), wherein, n* (n+1) is the Maximum edge numbers between n neighbor node, and m is viActual side between n neighbor node Number.Then, v is utilizediDegree of association μiAnd viTo neighbor node vjThe service diffusion success rate of information on services is spread, determines vjService is expanded Dissipate average speed, specifically, community network node viService average diffusion speedIts In, R represents dilation angle, and r is variable, 1≤r≤R, nrWhen expression dilation angle is r, radius is between r and radius (r-1) The number of neighbor node in annular region;Work as viFor source node when, αj rExpression radius be r- (r-1) region, node viSuccess To its neighbor node vjService the number of diffusion;βj rRepresent that radius is r to radius in the annular region of (r-1), node viTo its neighbor node vjService the total degree of diffusion.αj rj rRepresent viTo neighbor node vjSpread the success of information on services Rate,Represent viTo the success rate of all neighbor nodes diffusion information on services.Table Show in R radius regions and in 0~T time section interior nodes viService diffusion average speed.

Step S405:Calculate node viService diffusing capacity ψ(T)(vi)。

Utilize the service diffusion weightings wij T, service diffusion average speed Ui R, neighbor node quantity n, determine community network Node viService diffusing capacity

Step S406:By viDiffusing capacity is serviced compared with default diffusing capacity threshold value, if viDiffusing capacity is serviced more than default Diffusing capacity threshold value, then step S407 is performed, otherwise perform step S410.

Step S407:Work as viWhen servicing diffusing capacity more than default diffusing capacity threshold value, v is determinediFor advantage node, to viMake Advantage vertex ticks simultaneously preserves, i.e., by community network node viIt is used for in=(ID, NeighborList, Level, SList) Represent that the Level of community network node level is set to 1.

Step S408:Utilize advantage node vi, fast and effeciently spread trusted service.

Step S409:Determine viFor insincere node.

Step S410:Determine viFor ordinary node.

That is, service spread when, it is necessary to carry out node optimizing, that is, find ψ(T)(vi) the big node of value.Specifically Say, first determine whether community network node meets degree of belief threshold requirement (see the sentence 5 of following algorithm 1), if meeting to require, The then service average diffusion speed U of calculate nodei RDiffusing capacity weight w is serviced between nodeij T(see 6~language of sentence of algorithm 1 Sentence 25), and the service diffusing capacity ψ of calculate node(T)(vi), the advantage node to meeting threshold requirement make a mark and preserve (see 26~sentence of sentence 34 of algorithm 1).

Trusted service broadcast algorithm based on community network node optimizing is as follows:

Input:Social network service trusting relationship figure G, dilation angle R, specified time section T

Output:Set of node Target with advantage node label

BEGIN

Wherein, sentence 5 judges social network based on research of the early stage to credible community network node confidence and result of calculation Whether the degree of belief of network node meets threshold requirement.NeighborList in sentence 7 represents viNeighbor list.The present invention from Advantage node is searched out on the service diffusing capacity of node confidence and node, and advantage node has been done and marked (see sentence 29, mark Remember viLevel be 1, i.e. viTo meet the advantage node of service diffusing capacity threshold requirement), final obtain includes advantage node mark The set of node Target of will.

The time complexity of this method is calculated as follows:The size of the set of node of initial community network is M, to any one section Point, it averagely has | nr| individual neighbor node, it is R and in the case that run time section is T in dilation angle, algorithm complex is:O (M*R*|nr| * T)=O (N4)。

For any community network node v on service diffusion path pi, its state QiFor:{ini[1],…,ini[n], outi[1],…,outi[n] }, during node optimizing is carried out, the state variable in of nodej[i] and outi[j] 0, 1 } constantly change in value, be reflected as the dynamic process of service diffusion.For the first node on path, its original state is { 0 ..., 0,1 ..., 1 }, represent to prepare to carry out service diffusion to its neighbor node, the end-state variable of first node be 0 ..., 0,0 ..., 0 }, its service neighbor node that its path of warp-wise has been specified diffusion is represented.The original state of intermediate node and final shape State is:{ 0 ..., 0,0 ..., 0 }, play reception and transmission during diffusion is serviced, represent that receiving service spreads simultaneously Spread and service to other nodes.The original state of path ends node is { 0 ..., 0,0 ..., 0 }, and end-state for 0 ..., 0,1 ..., 1 }, represent to receive service diffusion but not yet spread to other nodes to service.

With reference to above-mentioned algorithm, parameter setting can be carried out according to table 1, and carry out emulation experiment.

The relative parameters setting of the emulation experiment of table 1

Performance evaluation includes two parts:Part I parser performance, that is, different parameters are analyzed to the present embodiment result Influence;Part II is the comparison and analysis in service diffusion to the present embodiment and classical model method.

The run time of the present embodiment expends as shown in figure 5, abscissa represents community network nodes, the section of community network Value of counting constantly is promoted within 50~1000 with step-length 50.viThe uniformly random generation in [1,1000].Can from Fig. 5 Know, the time performance of the present embodiment is continuously increased with the increase of community network nodes.Because the present embodiment is in the time Service diffusing capacity to community network trusted node is calculated in section T, and not the moment counted, in addition, in social network When network structure does not change or change is smaller, the statistical result of early stage can be calculated by the later stage to be utilized, and therefore, time performance is Feasible.

Performance of the present embodiment model (i.e. community network trusted service diffusion model) when trusted service spreads with it is following Several models are compared and analyzed:

(1) model (Flooding Model, FM) is flooded, the model does not require to safeguard the topological structure of network and correlation Router-level topology, require nothing more than and receive the node of information and carry out service diffusion in a broadcast manner;

(2) random walk model (Random Walk Model, RWM), the model is from community network node viNeighbours section Arbitrarily selection next node carries out service diffusion in point, if viNeighbor node have N number of, then taken by 1/N transition probability The diffusion of business, and service diffusion is carried out with specified dilation angle by random velocity.

Insincere node of the present invention, can not telecommunications services refer to that degree of belief does not reach the section of degree of belief threshold requirement Point, service.

Fig. 6 is for can not be in the case of telecommunications services, with the trusted service diffusivity ratio based on other models 50% or so Compared with schematic diagram, it will be appreciated from fig. 6 that in preferable the environment of social network, based on FM models than preferably realizing service diffusion, The model is not required to safeguard the node selection of the topological structure of network and correlation and calculated, and is required nothing more than and is received community network node Information on services is spread to its neighbor node with broadcast mode, until information on services be sent to destination node or the scope specified or Untill the life cycle of setting is limited to 0.But in actual the environment of social network often there is incredible community network node and Incredible service, according to FM models, it will cause flood attack and incredible information on services to be spread unchecked in community network, Such as in Fig. 6 when community network nodes are respectively 250,550,750, due to exist in community network 50% or so it is insincere Node and service, trusted service diffusivity is caused greatly to reduce, it is therefore, insincere when existing under society network environment When node and service, it is worthless to carry out servicing diffusion using flooding model.RWM models are by establishing community network node Which which transition probability matrix, determine to carry out service expansion to or the neighbor node of community network node actually by the matrix Dissipate, but for multifactor control problem, the model is difficult to global optimization and the trusted service of multifactor control satisfaction is maximum Diffusion.By contrast, the trusted service diffusion of information rate of this paper models is higher and relatively stable, test result indicates that this paper models There is preferable effect when solving trusted service and maximally diffusing problem.

The present invention based on community network interior joint to the trading situation between the calling situation and node of service, according to service Diffusion velocity, service diffusing capacity and confidence level, by community network node division into " advantage node " and " ordinary node ", wherein excellent Gesture node has the characteristics of service diffusion velocity is fast, service diffusing capacity is big, credible high.The simulation experiment result shows, of the invention Better performances, the target of expected setting can be reached, compared with decentralized control model (FM models and RWM models), in fact Existing trusted service spreads and service range of scatter is wider.

The hardware environment of this method emulation experiment can be:More Intel Duos i5,3.7GHz processors, 4GB internal memories Main frame etc., software environment can be:Operating system is Win7.Data set main source is:Carried out based on Epinions data sets Experiment, the data set, which is demonstrated by the community network node with trusting relationship and service of goods relation, the data set, to be included Trust-data and rating_data two parts, trust-data are the societies obtained from the Epinions community networks of reality Can network node and trust value therebetween.It is [0,1] to set the degree of belief span between node, trust of the node to itself Spend for 1.Ratings_data is evaluation information of the community network node to service of goods;The trust span of service for [0, 1].Because experiment needs, further optimization and supplement have been done to data set.

In summary, the present invention has following technique effect:

The present invention organically combines the service diffusion three of community network node trust, service trust and node, is based on Social network service trusting relationship, the optimal node (i.e. advantage node) of diffusion is serviced by finding, carries out the expansion of trusted service Dissipate, both considered network structure characteristic and propagation characteristic (including service diffusion velocity, service diffusing capacity), it is also considered that to social network Network node and the degree of belief of service, are solved in social network service diffusion process caused by insincere node and information on services Negative issue, improve the speed of service diffusion.

Although the present invention is described in detail above, the invention is not restricted to this, those skilled in the art of the present technique Various modifications can be carried out according to the principle of the present invention.Therefore, all modifications made according to the principle of the invention, all should be understood to Fall into protection scope of the present invention.

Claims (10)

  1. A kind of 1. trusted service method of diffusion based on community network node optimizing, it is characterised in that including:
    Success rate is spread according to community network node serve, it is determined that the degree of belief of each community network node;
    According to the degree of belief of each community network node, it is determined that whether each community network node is trusted node;
    For identified trusted node, diffusion number of success and success rate, trusted node and neighbours are serviced according to trusted node Associated volume, the neighbor node quantity of node, determine the service diffusing capacity of trusted node;
    Diffusing capacity is serviced according to trusted node, determines whether trusted node is advantage node, so as to advantage section determined by utilizing Point carries out trusted service diffusion.
  2. 2. according to the method for claim 1, it is characterised in that determine some community network node v by following stepsj Degree of belief:
    According to neighbor node viTo vjThe service diffusion success rate of diffusion service, determines viTo vjDegree of belief;
    Utilize identified viTo vjDegree of belief and viWeight, determine vjDegree of belief.
  3. 3. according to the method for claim 2, it is characterised in that determine v by following stepsjWhether it is trusted node:
    By vjDegree of belief compared with default degree of belief threshold value;
    Work as vjDegree of belief when being more than default degree of belief threshold value, determine vjFor trusted node, v is otherwise determinedjFor insincere node.
  4. 4. according to the method for claim 3, it is characterised in that when it is determined that vjFor trusted node when, determined by following steps vjService diffusing capacity:
    According to vjTo viThe service diffusion number of success of diffusion service, determines vjTo viSpread the service diffusion weightings of service;
    According to vjWith viAssociated volume, determine vjThe degree of association;
    Utilize vjTo viThe service diffusion success rate of diffusion service, vjThe degree of association, determine vjService diffusion average speed;
    Using the service diffusion weightings, service diffusion average speed, neighbor node quantity, v is determinedjService diffusing capacity.
  5. 5. according to the method described in claim any one of 2-4, it is characterised in that determine v by following stepsjWhether it is advantage Node:
    By vjDiffusing capacity is serviced compared with default diffusing capacity threshold value;
    Work as vjWhen servicing diffusing capacity more than default diffusing capacity threshold value, v is determinedjFor advantage node, v is otherwise determinedjFor ordinary node.
  6. A kind of 6. trusted service disperser based on community network node optimizing, it is characterised in that including:
    Degree of belief determining module, for spreading success rate according to community network node serve, it is determined that each community network node Degree of belief;
    Trusted node determining module, for the degree of belief according to each community network node, it is determined that each community network node is No is trusted node;
    Diffusing capacity determining module, for identified trusted node, according to trusted node service diffusion number of success and success rate, Trusted node and the associated volume of neighbor node, neighbor node quantity, determine the service diffusing capacity of trusted node;
    Advantage node determining module, for servicing diffusing capacity according to trusted node, determine whether trusted node is advantage node, with Just advantage node carries out trusted service diffusion determined by.
  7. 7. device according to claim 6, it is characterised in that for some community network node vj, the degree of belief is true Cover half root tuber is according to neighbor node viTo vjThe service diffusion success rate of diffusion service, determines viTo vjDegree of belief, and using institute really Fixed viTo vjDegree of belief and viWeight, determine vjDegree of belief.
  8. 8. device according to claim 7, it is characterised in that the trusted node determining module is by vjDegree of belief with it is pre- If degree of belief threshold value is compared, work as vjDegree of belief when being more than default degree of belief threshold value, determine vjFor trusted node, otherwise really Determine vjFor insincere node.
  9. 9. device according to claim 8, it is characterised in that when it is determined that vjFor trusted node when, the diffusing capacity determines mould Root tuber is according to vjTo viThe service diffusion number of success of diffusion service, determines vjTo viSpread the service diffusion weightings of service;According to vj With viAssociated volume, determine vjThe degree of association;Utilize vjTo viThe service diffusion success rate of diffusion service, vjThe degree of association, it is determined that vjService diffusion average speed, and using the service diffusion weightings, service diffusion average speed, neighbor node quantity, determine vj Service diffusing capacity.
  10. 10. according to the device described in claim any one of 7-9, it is characterised in that the advantage node determining module is by vjService Diffusing capacity works as v compared with default diffusing capacity threshold valuejWhen servicing diffusing capacity more than default diffusing capacity threshold value, v is determinedjFor advantage Node, otherwise determine vjFor ordinary node.
CN201510023305.9A 2015-01-16 2015-01-16 A kind of trusted service method of diffusion and device based on community network node optimizing CN104539731B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201510023305.9A CN104539731B (en) 2015-01-16 2015-01-16 A kind of trusted service method of diffusion and device based on community network node optimizing

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201510023305.9A CN104539731B (en) 2015-01-16 2015-01-16 A kind of trusted service method of diffusion and device based on community network node optimizing

Publications (2)

Publication Number Publication Date
CN104539731A CN104539731A (en) 2015-04-22
CN104539731B true CN104539731B (en) 2017-12-26

Family

ID=52855198

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201510023305.9A CN104539731B (en) 2015-01-16 2015-01-16 A kind of trusted service method of diffusion and device based on community network node optimizing

Country Status (1)

Country Link
CN (1) CN104539731B (en)

Families Citing this family (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN107342975B (en) * 2016-12-21 2020-03-24 安徽师范大学 Domain division-based trust computing method in untrusted cloud environment
CN110577126B (en) * 2019-09-23 2020-06-02 上海大学 Intelligent elevator safety management system

Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102546230A (en) * 2010-12-08 2012-07-04 中国科学院声学研究所 Overlay-network topological optimization method in P2P (Peer-To-Peer) streaming media system
CN103888373A (en) * 2014-03-05 2014-06-25 北京邮电大学 Adjusting method and device of save file storage capacity of network nodes

Family Cites Families (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US8503334B2 (en) * 2007-12-14 2013-08-06 Level 3 Communications, Llc System and method for providing network services over shared virtual private network (VPN)

Patent Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102546230A (en) * 2010-12-08 2012-07-04 中国科学院声学研究所 Overlay-network topological optimization method in P2P (Peer-To-Peer) streaming media system
CN103888373A (en) * 2014-03-05 2014-06-25 北京邮电大学 Adjusting method and device of save file storage capacity of network nodes

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
基于社会网络动态反馈的Web服务信任度计算;张佩云 等;《模式识别与人工智能》;20130430;全文 *

Also Published As

Publication number Publication date
CN104539731A (en) 2015-04-22

Similar Documents

Publication Publication Date Title
Wang et al. Wireless big data computing in smart grid
Jesus et al. A survey of distributed data aggregation algorithms
Kazemi et al. Geotrucrowd: trustworthy query answering with spatial crowdsourcing
de Assuncao et al. Distributed data stream processing and edge computing: A survey on resource elasticity and future directions
US9660876B2 (en) Collector mechanisms in a content delivery network
Su et al. Steiner tree based optimal resource caching scheme in fog computing
Naranjo et al. FOCAN: A Fog-supported smart city network architecture for management of applications in the Internet of Everything environments
US9426233B2 (en) Multi-objective server placement determination
Hegeman et al. Toward optimal bounds in the congested clique: Graph connectivity and MST
Zhang et al. A unified enhanced particle swarm optimization‐based virtual network embedding algorithm
Bandara et al. Collaborative applications over peer-to-peer systems–challenges and solutions
US9680919B2 (en) Intelligent messaging grid for big data ingestion and/or associated methods
US9098344B2 (en) Cloud-edge topologies
Abel et al. Cross-system user modeling and personalization on the social web
JP4750457B2 (en) Data overlay, self-organizing metadata overlay and application level multicasting
Cohen et al. Almost optimal virtual machine placement for traffic intense data centers
Hu et al. Flutter: Scheduling tasks closer to data across geo-distributed datacenters
Riva et al. The urbanet revolution: Sensor power to the people!
Al Ridhawi et al. A collaborative mobile edge computing and user solution for service composition in 5G systems
US8200587B2 (en) Techniques to filter media content based on entity reputation
US20140156360A1 (en) Dynamic expressions for representing features in an online system
Lenzen et al. Efficient distributed source detection with limited bandwidth
Doulkeridis et al. Peer-to-peer similarity search in metric spaces
Zhang et al. Virtual network embedding based on the degree and clustering coefficient information
Li et al. Efficient range query processing in peer-to-peer systems

Legal Events

Date Code Title Description
PB01 Publication
C06 Publication
C10 Entry into substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
CF01 Termination of patent right due to non-payment of annual fee

Granted publication date: 20171226

Termination date: 20190116

CF01 Termination of patent right due to non-payment of annual fee