CN103701939B - Method for interchanging data - Google Patents
Method for interchanging data Download PDFInfo
- Publication number
- CN103701939B CN103701939B CN201410018024.XA CN201410018024A CN103701939B CN 103701939 B CN103701939 B CN 103701939B CN 201410018024 A CN201410018024 A CN 201410018024A CN 103701939 B CN103701939 B CN 103701939B
- Authority
- CN
- China
- Prior art keywords
- node
- data
- trust
- evaluating data
- evaluating
- 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
Links
Landscapes
- Computer And Data Communications (AREA)
- Data Exchanges In Wide-Area Networks (AREA)
Abstract
The invention provides data interactive method, including:Feedback node will be to that will carry out the trust vector set of multiple Section Points of data interaction in first node acquisition network in network;First node determines the trust value of multiple Section Points according to trust vector set;First node determines to carry out the interactive object node of data interaction according to trust value from multiple Section Points;First node carries out data interaction with the interactive object node determined.Present invention also offers a kind of data interaction system.The present invention can be effectively reduced erroneous judgement number of times, so as to promote the reliabilty and availability that network node data is interacted.
Description
Technical field
The present invention relates to network information processing technology, more particularly to a kind of data interactive method in a peer-to-peer network.
Background technology
In peer to peer environment, it is frequently necessary to carry out data interaction between the main body do not known each other mutually to complete a certain work
It is dynamic, network trading, pp file transmission are downloaded etc..Data volume is very huge in network, and true and false data are mixed
It is miscellaneous.Unworthy or false data are how filtered out from mass data, obtains and utilizes valuable True Data, for right
Effective utilize Deng network is critically important.
By taking the service transacting of peer to peer environment as an example, the subject identity of service can be divided into service requester and be carried with service
Donor.When a requestor faces multiple alternative ISPs, requestor how to determine one it is credible, reliable
ISP simultaneously obtains quality services to it, is the key technical problem in peer-to-peer network.
Service evaluation data for ISP are to determine that one of the credible and degree of reliability of ISP sentences
Disconnected foundation.But, social virtual and non-direct contact of complexity, network etc. so that for commenting for ISP
Valence mumber according to the evaluating data that there are authentic assessment data, incomplete objective appraisal data and false malice (hereinafter referred to as
" unreal evaluating data ") etc. various situations.If unreal evaluating data of leaving is present, then people can not just distinguish or confirm clothes
The real service quality of business supplier, causes confusion and the destruction of network trading and service environment.Therefore, in network service ring
In border, it is necessary to which unreal evaluating data is filtered.
At present on the problem of unreal evaluating data is filtered, the filter method based on irrelevance is employed mostly, such side
The main thought of method is:With evaluating data difference degree of the evaluating data person to same ISP, go to portray irrelevance
Size, it is believed that the big node of irrelevance is exactly malicious node.This hypothesis is not inconsistent with actual conditions, easily causes erroneous judgement.It is true
On, node is directly simply divided to honest node and malicious node into, it is impossible to summarize the behavior of node in systems in practice completely special
Levy, erroneous judgement can be caused.For example:If the service that certain node is provided is an e-book, some users like the book, some users
But do not like, this species diversity is the subjectivity generation by user mostly;But if it is a disease that certain node, which slanders the e-book,
Malicious file, is commented to going on business, and this is likely to malice evaluating data.Therefore subjective assessment data and malice evaluating data are very likely
Uniformity can be shown so that both evaluating data similarities are very big, cause algorithm to judge by accident.Equally, part of nodes is to certain
Interaction is reluctant to provide valuable evaluating data, selects the evaluating data of system default, show passivity.It is this to evaluate
Data, with malicious node formation advantage collusion, can also make in itself and without malicious, but during evaluating data is filtered
Algorithm failure based on evaluating data irrelevance.Therefore, the effect of existing unreal evaluating data filter method is unsatisfactory.
The content of the invention
It is an object of the invention to provide method for interchanging data, the validity of data interaction is improved.
According to an aspect of the invention, there is provided data interactive method, including:First node in network obtains network
Middle feedback node will be to that will carry out the trust vector set of multiple Section Points of data interaction;First node is according to trust vector
Set determines the trust value of multiple Section Points;First node is according to the trust value of determination, and being determined from multiple Section Points will
Carry out the interactive object node of data interaction;First node carries out data interaction with the interactive object node determined.
In some embodiments, wherein trust vector set according to feedback node to the evaluating data of each Section Point to
Quantity setAnd the feedback degree of belief of each feedback node is determined, whereinBe feedback node it
One riTo Section Point pjEvaluating data vector;And/or wherein determine that the trust value of Section Point includes:To trust vector collection
Each node r in conjunctioni, selection reservation confidence level higher evaluating data and corresponding node;With the corresponding section according to selection
The evaluating data of point, determines the trust value of Section Point.
In some embodiments, the step of selection retains confidence level higher evaluating data and corresponding node, can wrap
Include:Recognize each node in trust vector setWhether insecure evaluating data is included;With deletion trust vector collection
The corresponding node of unreliable evaluating data in conjunction.
In above-mentioned steps, it may include according to the evaluating data of the node of selection, the step of the trust value for determining Section Point:
According toExceptional value threshold range is set, the evaluation for being higher than anomaly evaluation high thresholdhigh in set is deleted
Data and the evaluating data less than anomaly evaluation threshold value lower limit;To the set after suppressing exception thresholding, extract any two groups and comment
Valence mumber evidence makes the difference and is grouped;For the evaluating data of packet, the trust evaluation set in each node to Section Point is selected
Proportion is more in TrustTable one group;One group of evaluating data with to selection, takes weighted average, obtains feedback and trusts
Spend Γf, direct degree of belief Γ is obtained from TrustTable inquiriesd, according to direct degree of belief ΓdWith feedback degree of belief Γf, it is determined that
The trust value trust of Section Pointj。
In some embodiments, it can also include:Determine whether the trust value trust of Section PointjWhether it is more than
Predetermined value, if it is judged that being more than predetermined value for trust value, then first node determines that the Section Point is interactive object node,
And carry out data interaction with interactive object node.
In some embodiments, degree of belief determining unit is according to evaluating data vector of the feedback node to each Section Point
CollectionAnd the feedback degree of belief of each feedback node is determined, whereinIt is one of feedback node
riTo Section Point pjEvaluating data vector.Wherein, degree of belief determining unit can also be to each node in trust vector set
ri, selection reservation confidence level higher evaluating data and corresponding node;With the evaluating data of the respective nodes according to selection, really
Determine the trust value of Section Point.
In some embodiments, degree of belief determining unit can recognize that each node in trust vector setIt is
It is no to include insecure evaluating data, and delete the corresponding node of unreliable evaluating data in trust vector set.
In some embodiments, degree of belief determining unit also performs following operation:
According toExceptional value threshold range is set, the evaluation for being higher than anomaly evaluation high thresholdhigh in set is deleted
Data and the evaluating data less than anomaly evaluation threshold value lower limit;To the set after suppressing exception thresholding, extract any two groups and comment
Valence mumber evidence makes the difference and is grouped;For the evaluating data of packet, the trust evaluation set in each node to Section Point is selected
Proportion is more in TrustTable one group;One group of evaluating data with to selection, takes weighted average, obtains feedback and trusts
Spend Γf, direct degree of belief Γ is obtained from TrustTable inquiriesd, according to direct degree of belief ΓdWith feedback degree of belief Γf, it is determined that
The trust value trust of Section Pointj。
In some embodiments, data interaction unit can determine whether the trust value trust of Section PointjWhether it is more than predetermined
Value, if it is judged that being more than predetermined value for trust value, then first node determines that the Section Point is interactive object node, and with
Interactive object node carries out data interaction.
The present invention can be effectively reduced erroneous judgement number of times when data interaction is carried out between network node, so as to promote network section
The reliabilty and availability of point data interaction.
Brief description of the drawings
Fig. 1 is the data interaction architectural schematic of one embodiment of the present invention;
Fig. 2 is the method flow schematic diagram of the data interaction of one embodiment of the present invention.
Fig. 3 is the data interaction system schematic diagram of one embodiment of the present invention.
Embodiment
Below in conjunction with the accompanying drawings and specific embodiment the present invention is described in further detail.
As shown in figure 1, the data interactive method that the present invention is provided can be applied in peer to peer environment.Hereinafter,
Service requester is as first node 100, and ISP, as first node 100, is service requester (first node 100)
It is feedback node 200 to provide to ISP (first node 100) evaluating data feedback.It is appreciated that first node
100th, the node type of first node 100 and feedback node 200 be not it is fixed, can according in data interaction perhaps its
Its factor alternative types.
The peer to peer environment does not have core super node, thus evaluating data is not about ISP's expansion.
During service requester obtains service, service requester active collection Feedback Evaluation data, to the Feedback Evaluation being collected into
Data are filtered and recognized, and then the degree of belief of ISP is estimated, and to degree of belief, high ISP saves
Point obtains service, and makes evaluating data to the actual interaction of this service.Meanwhile, dynamic is updated to ISP's
Trusted status, provides trusted status foundation, TrustTable includes trust vector, service request for forwarding next time TrustTable
The trust estimation that person, ISP and service requester are made to ISP.
The generation and forwarding of Feedback Evaluation data are carried out as follows.Feedback Evaluation data source is in Feedbacks
Peers, Feedbacks Peers are provided clothes to be not involved in other nodes of certain interaction in network into interaction node
Business node (node for asking service) feedback dependent credit message, mainly trust vector.Feedbacks Peers are to service
The evaluating data vector set of supplier is handled, and wherein evaluating data vector representation isFollowing article
It is described, calculated by evaluating data vector and obtain trust vector.Evaluating data vector is stored in LocalHistoryArray, bag
R containing service requesteri, ISP pj, and riTo pjThe evaluating data ratings that the service of offer is provided.
It is assumed that node riTo pjSet D={ the d of the evaluating data provided in h times nearest interaction1,d2,…,dh, its
Middle dk∈[0,1],k∈[1,h],h≤10.Element in set D derives from LocalHistoryArray, arranges in chronological order
Row, d1It is from the currently more long evaluation that once interaction is produced, dhRepresent the evaluation produced from current recent interaction.Represent node riTo pjElement in the evaluating data set D provided in h times nearest interaction is merged
Evaluating data.Fusion is carried out as follows:
With reference to the cognitive process of psychology of human society, it should can most reflect present node pair from current recent interaction
Receive the authentic assessment data cases of service.Therefore, more weights, the relatively early interaction warp of reduction are given to the interbehavior of kainogenesis
The weight tested, the historical factors γ (k), γ (k) ∈ [0,1] decayed in superior one of evaluating data with the time.Wherein h is riIt is right
Same node pjThe number of evaluating data history vectors.Historical factors γ (k) calculation formula are as follows:
In peer to peer environment, status is equal between node, does not act as third party's node of agency, also just not credible
Bad feedback person, broadcast feedback information is unique selection.Each node forwards oneself by the way of fixed time broadcast to trusted neighbor
TrustTable record.
In order to which the degree of belief to ISP is estimated, the degree of belief for calculating ISP includes 2 parts:Directly
Connect degree of belief and feedback degree of belief, by by direct degree of belief with feed back both degree of beliefs result be superimposed estimating for degree of establishing trust
Survey.Direct degree of belief is the direct trust evaluation data of ISP, and feedback degree of belief provides for service requester with service
After person's interaction data, the trust evaluation data that service requester feeds back to ISP.
The calculating of direct degree of belief:It is node riTo pjThe evaluation number provided in h times nearest interaction
According to the fusion of collection.As described above.
Feed back the calculating of degree of belief:In trust vector TrustTable setIn,
Wherein NuThe passive evaluating data after (see below explanation), u ∈ [1, t], S are filtered for ratingsyIt is for after ratings filterings
Subjective assessment data, y ∈ [1, q], BvIt is the malice evaluating data after being filtered for ratings, v ∈ [1, m], HeIt is for ratings
Honest evaluating data after filtering, e ∈ [1, n].Trust vector is weighted and averagely obtains feeding back degree of belief.Wherein recognize
Passive evaluating data, malice evaluating data is corrupt data, and weight is 0, and subjective assessment data weighting is 0.5, and honesty is commented
Valency data weighting is 1.Weighted average calculation formula is as follows:
The estimation of degree of belief:It is by the way that direct degree of belief is superimposed with feeding back both degree of beliefs, according to formula trustj=w1
×Γd+w2×ΓfOperation, w1And w2It is respective weight respectively, meets w1+w2=1,
w1∈[0,1],w2∈[0,1]。(w1And w2It is that w is used in empirical parameter, this example1=w2=0.5)
Dynamic updates trusted status, and node completes the assessment and prediction of degree of belief according to mathematical modeling, dynamically believed
Appoint recalculating for value.Before node A and node B interactions (such as one time service transacting), the node A of evaluating data main body is used as
Estimate the trust value of the node B as evaluating data object.If trustj< 0.5, is considered as not physical node, then reselects
Evaluating data object.Otherwise, interaction can be completed.After interaction is completed, node A needs to update in TrustTable to B's
Trust value.
If this Feedback Evaluation data of interaction based on feedback person is as judgement basis, node A updates it to node B's
The credential state value for the main body that feedback is provided it is inquired about during trust value in TrustTable, if sincere node then receives instead
The Feedback Evaluation of feedback person, otherwise refuses the Feedback Evaluation of feedback person.
The trust messages that the present invention feeds back are obtained by broadcasting.If not physical node around node, refusal was from should
The Feedback Evaluation data of neighbor node, i.e., to the not physical node identified, change node trusted status.Specifically, A pairs of node
The trust state of oneself neighbor node carries out filter analysis, if oneself neighbor node is not physical node, its trust state is set
Be set to it is insincere, and refuse after the Feedback Evaluation data from the neighbor node.
TrustTable is to adjacent trusted node for node timing forwarding, is commented while only receiving the feedback that trusted node provides
Valence mumber evidence.If receiving the feedback of neighbor node offer indiscriminately, it is easy to because neighbor node is not physical node, receive
The feedback of the offer of physical node is not real data yet, so as to easily produce refusal trust service or receive the feelings of bad service
Condition.Thus node trusted status has very strong actual use value.
As shown in Fig. 2 data filtering method according to an embodiment of the present invention includes:
Node A searches for obtain service (step S201), wherein nodes set ρ={ p from network1,p2,…
pnA certain service can be provided to node A.Node A selects egress p from set ρj(1≤j≤n) (step S202).
It is p that node A extracts ISP from TrustTablejTrust vector set φ.The trust vector collection
Close each vector in φRepresent riNode is to pjThe evaluating data that provides of service that node is provided is(S203)。
To the r in the trust vector set φ that is extracted in step S203i, recognize that it is provided no for passive evaluating data
(step S204).If riTo arbitrary node p in TrustTablejThe evaluating data provided all, then judges to evaluate number
According to being passive evaluating data, node riFor passive node;Otherwise, next r is carried outiThe judgement of node.So continue, until
Last riNode judges to finish.If each node riThe evaluation number that arbitrary other nodes are provided in TrustTable
According to all different, then it is assumed that without passiveness evaluating data in trust vector set φ evaluating data, also without passiveness in set
Node.
If in step S204, the evaluating data in set φWhen being judged as non-passive evaluating data, enter one
Step judges remaining each evaluating dataValue;Otherwise by the evaluating data in set φ in delete step S204
It is judged as the node r of passive evaluating datai, further judge other evaluating datasValue.In experimentation, withThe scope concentrated of value as range of normal value, by checking, the scope is set to 80%, contributes to these different
Which block the filtering of constant value, can illustrate whole evaluation set concentrates on, and is so easy to find out abnormity point, meanwhile, it will not miss again
Delete other evaluations.
If wherein 80%Value is both greater than equal to 0.5, then evaluating data exceptional value thresholding is set into first
Threshold range, such as 0.9 and 0.2;If wherein 80%Value both less than 0.5, then set Second Threshold scope, example
Such as 0.8 and 0.1.
According to step S204 setting, delete in set φ higher than the anomaly evaluation data gate limit value upper limit evaluating data with
Less than the evaluating data of anomaly evaluation data gate limit value lower limitSo as to delete abnormal evaluating data (S205).Stay
Under evaluating data may include part subjective assessment data, honest evaluating data and malice evaluating data.Because not being
All subjective nodes be all it is abnormal, it is most of or honest, thus by given threshold scope so as to identifying subjectivity
Node.The evaluating data that subjective node is provided, can be limited by subjective factors such as itself node knowledges, to same clothes
Business, might have larger difference.Therefore, in this example, extra high evaluating data and especially low evaluating data are all picked
Except going out, distinguished point is deleted, contributes to the follow-up processing to data, the calculating of average value will not be influenceed because of distinguished point,
So as to reach the purpose of identification.
Evaluating data in the set that step S205 is determined includes honest evaluating data, and malice evaluating data, part is main
See evaluating data.All subjective assessment data are not deleted, those abnormal nodes why are deleted, are because above-mentioned exception
Node can cause very big influence for the selection of standard.
Any two evaluating data is extracted to make the difference.Two evaluating datas (a, b), it is assumed that wherein a is standard, makes the difference and obtains one
Number Sab.An evaluating data c is extracted from evaluating data set again, is made the difference with standard a and is compared Sac, if Sac<Sab, then a
With c in a set, otherwise c and b is in a set;An evaluating data d is extracted from evaluating data set again, Sad is made the difference to obtain,
If Sad<Min { Sac, Sab }, if Sac<Sab, then d and c is in a set, if Sac>Sab, then d and b is in a set
It is interior;If Sad>Max { Sac, Sab }, if Sac<Sab, then d and b is in a set, if Sac>Sab, then d and c is in a collection
In closing;Whom if Sad is situated between, judge, closer to, whom to belong to a set together with accordingly.So until all extractions
Complete evaluating data.
Two groups of evaluating datas being so divided into, the difference of corresponding evaluating data can meet the difference of each group desk evaluation data
Value tends to 0.Because the number if over evaluation increases, same class service requester should be derived from by evaluating set with group, right
The evaluation that same ISP provides only exists random error, thus is that the difference that can meet each group desk evaluation data tends to
In 0, if the difference for meeting each group desk evaluation data is intended to 0, illustrate in two groups of evaluation set being divided into, each group of evaluation
The service evaluation person of same generic attribute ((sincerity) with a high credibility or (malice) with a low credibility) is all derived from, honesty is realized and comments
Valency and malice evaluate the classification of two class data, are divided into A and B groups (step S206).
To step S206 A and B group evaluating datas, it is big that selection retains the evaluation provided proportion in TrustTable
One group of data as believable node (service requester) evaluation (S207).The one group of evaluating data retained step S207
Handled to obtain feedback degree of belief Γf, direct degree of belief ΓdObtained from TrustTable inquiries, estimation egress pjLetter
Appoint value trustj(step S208), the estimation step of for example above-mentioned degree of belief of computational methods.
If the node p of estimationjTrust value trustj>=0.5, then node A and pjStart transaction.Otherwise, reselect
Node, returns to step S202 (step S209).
Node A and node pjComplete after transaction, provide evaluating data, preserve evaluating data (step S210).Again to pjInstead
Present node 200 and carry out trusted status classification, overall process terminates.
42 nodes are provided with the analogue system of example, by honest evaluating data:Subjective assessment data:Malice is evaluated
Data:Passive evaluating data is 7:7:4:3 ratio generation, each node independently completes interactive history at random.In the evaluation of generation
In data basis, a pair of nodes (A, B) are selected at random, and its interior joint A estimation nodes B trust value is separately operable present invention side
Method, the filter method based on irrelevance and calculate trust value without filter algorithm, irrelevance refer between evaluating data away from
From for inventive algorithm, no filter algorithm refers to:A pair of the nodes selected at random, number is evaluated without the method filtering of the present invention
According to directly initiation request calculates False Rate, using False Rate as performance indications.False Rate is above-mentioned sides of the node A by the present invention
The trust value that method is determined mistakenly allows node A to initiate to ask the proportion shared by the number of times of service to the node B for providing difference service.Examination
Test and carried out 10 times, experimental condition altogether:It is required that above-mentioned node is generated by corresponding proportion, ratio refers to node ratio, evaluates number
According to generation be random, each node independently completes interactive history at random, that is to say, that the generation of evaluating data be it is random, this
Sample can guarantee that each node equiprobability of experiment participates in interaction.Enhancing is credible, and node is generated in proportion, and autonomous random completion interaction is gone through
History, filters evaluating data.1000 wheels are done accordingly, False Rate is counted, said process is repeated 10 times, averages, obtains
Final False Rate 5.1%, 20.4%, 50.9% illustrates that innovatory algorithm can effectively reduce erroneous judgement number of times, so as to promote
The reliabilty and availability of system.
According to another aspect of the present invention there is provided a kind of data interaction system, include as shown in Figure 3:First in network
Node 100, the multiple feedback nodes 200 that can be communicated with first node 100, and can be with first node 100 and feedback node 200
The Section Point 300 of data interaction.Wherein, first node 100 includes degree of belief determining unit 101, obtains and section is fed back in network
200 pairs of point will carry out the trust vector set of multiple Section Points 300 of data interaction, be determined according to trust vector set many
The trust value of individual Section Point 300;With data interactive unit 102, according to the trust value of determination, from multiple Section Points 300
It is determined that to carry out the interactive object node of data interaction, and data interaction is carried out with the interactive object node of determination.
Degree of belief determining unit 101 is according to the evaluating data vector sets of 200 pairs of each Section Points 300 of feedback nodeAnd the feedback degree of belief of each feedback node 200 is determined, whereinIt is feedback node 200
One of riTo Section Point 300pjEvaluating data vector.Wherein, degree of belief determining unit 101 can also be in trust vector set
Each node ri, selection reservation confidence level higher evaluating data and corresponding node;With the respective nodes according to selection
Evaluating data, determines the trust value of Section Point 300.Determine the method for the trust value of Section Point 300 as described above.
In addition, degree of belief determining unit 101 can recognize that each node in trust vector setWhether include can not
The evaluating data leaned on, and delete the corresponding node of unreliable evaluating data in trust vector set.Identification method is as described above.
Each node in identification trust vector set is performed of degree of belief determining unit 101Whether include can not
The evaluating data police leaned on, can perform following operate:
According toExceptional value threshold range is set, the evaluation for being higher than anomaly evaluation high thresholdhigh in set is deleted
Data and the evaluating data less than anomaly evaluation threshold value lower limit;To the set after suppressing exception thresholding, extract any two groups and comment
Valence mumber evidence makes the difference and is grouped;For the evaluating data of packet, the trust evaluation set in each node to Section Point 300 is selected
Proportion is more in TrustTable one group;One group of evaluating data with to selection, takes weighted average, obtains feedback and trusts
Spend Γf, direct degree of belief Γ is obtained from TrustTable inquiriesd, according to direct degree of belief ΓdWith feedback degree of belief Γf, it is determined that
The trust value trust of Section Point 300j。
Data interaction unit 102 can determine whether the trust value trust of Section Point 300jWhether predetermined value is more than, if it is determined that
As a result it is more than predetermined value for trust value, then first node 100 determines that the Section Point 300 is interactive object node, and with interacting
Object node carries out data interaction.
Data interaction system also includes trusting table data cell 103, letter of the storage first node 100 to Section Point 300
Appoint value, and update after data interaction the trust value to the Section Point 300 of the data interaction, and trust value is sent to first
Node 100 and feedback node 200.
The present invention can be effectively reduced erroneous judgement number of times when data interaction is carried out between network node, so as to promote network section
The reliabilty and availability of point data interaction.
Claims (5)
1. data interactive method, including:
Feedback node (200) will be to that will carry out multiple the second of data interaction in first node (100) acquisition network in network
The trust vector set of node (300);
The first node (100) determines the trust value of the multiple Section Point (300) according to the trust vector set;
The first node (100) determines to carry out data friendship according to the trust value from the multiple Section Point (300)
Mutual interactive object node;
The first node (100) and the interactive object node of the determination carry out data interaction;
The trust vector set is according to evaluating data vector of the feedback node (200) to each Section Point (300)
CollectionAnd the feedback degree of belief of each feedback node (200) is determined, whereinIt is described
One of feedback node (200) riTo the Section Point (300) pjEvaluating data vector;
The trust value for determining the Section Point (300) includes:
To each node r in the trust vector seti, select to retain the higher evaluating data of confidence level and corresponding
Node;
According to the evaluating data of the respective nodes of the selection, the trust value of the Section Point (300) is determined;
The step of selection retains confidence level higher evaluating data and corresponding node, includes:
Recognize each node in the trust vector setWhether insecure evaluating data is included;
Delete the corresponding node of unreliable evaluating data in the trust vector set;
The step of evaluating data of the node according to the selection, trust value for determining the Section Point (300), includes:
According toSet exceptional value threshold range, delete set in higher than anomaly evaluation high thresholdhigh evaluating data and
Less than the evaluating data of anomaly evaluation threshold value lower limit;
Delete in set higher than the evaluating data of anomaly evaluation high thresholdhigh and less than anomaly evaluation threshold value lower limit described
Evaluating data the set, extract any two groups of evaluating datas and make the difference and be grouped;
For the evaluating data of the packet, the trust evaluation set in each node to the Section Point (300) is selected
Proportion is more in TrustTable one group;
To one group of evaluating data of the selection, weighted average is taken, the feedback degree of belief Γ is obtainedf, from the TrustTable
Inquiry obtains direct degree of belief Γd, according to the direct degree of belief ΓdWith feedback degree of belief Γf, determine the Section Point
(300) trust value trustj;
Determine whether the trust value trust of the Section Point (300)jWhether predetermined value is more than;
If it is judged that being more than predetermined value for the trust value, then the first node (100) determines the Section Point (300)
For the interactive object node, and data interaction is carried out with the interactive object node.
2. data interactive method according to claim 1, wherein the feedback letter appoints degree to obtain in the following manner:
In the trust vector set TrustTable of the feedback node (200), if
In, wherein NuFor passive evaluating data, u ∈ [1, t], SyIt is for subjective evaluating data, y ∈ [1, q], BvIt is malice evaluating data,
V ∈ [1, m], HeIt is honest evaluating data, e ∈ [1, n] are then described to feed back degree of belief ΓfBy to the passive evaluating data,
The weighted average of malice evaluating data, subjective assessment data and honest evaluating data is obtained.
3. data interactive method according to claim 2, wherein the ΓfWeighted average mode be:
。
4. data interactive method according to claim 3, wherein the Section Point (300) degree of belief trustjAccording to such as
Under type is obtained:
trustj=w1×Γd+w2×Γf
Wherein, w1And w2It is weight coefficient respectively, meets w1+w2=1, w1∈[0,1],w2∈[0,1]。
5. the data interactive method according to claim any one of 1-4, in addition to:The first node (100) and described
Interactive object node is completed after data interaction, generates the evaluating data to the interactive object node.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201410018024.XA CN103701939B (en) | 2014-01-16 | 2014-01-16 | Method for interchanging data |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201410018024.XA CN103701939B (en) | 2014-01-16 | 2014-01-16 | Method for interchanging data |
Publications (2)
Publication Number | Publication Date |
---|---|
CN103701939A CN103701939A (en) | 2014-04-02 |
CN103701939B true CN103701939B (en) | 2017-07-28 |
Family
ID=50363333
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201410018024.XA Active CN103701939B (en) | 2014-01-16 | 2014-01-16 | Method for interchanging data |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN103701939B (en) |
Families Citing this family (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN106911660B (en) * | 2016-08-02 | 2020-12-08 | 中国移动通信集团设计院有限公司 | Information management method and device |
CN111212139A (en) * | 2020-01-02 | 2020-05-29 | 支付宝(杭州)信息技术有限公司 | Method and device for updating trust node information |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101714976A (en) * | 2009-10-15 | 2010-05-26 | 浙江大学 | Method for resisting malicious behaviors of nodes in P2P network |
CN102244587A (en) * | 2011-07-15 | 2011-11-16 | 杭州信雅达数码科技有限公司 | Method for trust evaluation of nodes in network |
CN103237023A (en) * | 2013-04-16 | 2013-08-07 | 安徽师范大学 | Dynamic trust model establishing system |
-
2014
- 2014-01-16 CN CN201410018024.XA patent/CN103701939B/en active Active
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101714976A (en) * | 2009-10-15 | 2010-05-26 | 浙江大学 | Method for resisting malicious behaviors of nodes in P2P network |
CN102244587A (en) * | 2011-07-15 | 2011-11-16 | 杭州信雅达数码科技有限公司 | Method for trust evaluation of nodes in network |
CN103237023A (en) * | 2013-04-16 | 2013-08-07 | 安徽师范大学 | Dynamic trust model establishing system |
Non-Patent Citations (1)
Title |
---|
一种新的无线传感器网络节点信誉评测模型AEMP;顾翔等;《计算机科学》;20111215;第38卷(第12期);第92-95页 * |
Also Published As
Publication number | Publication date |
---|---|
CN103701939A (en) | 2014-04-02 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Liu et al. | Large-scale group decision making model based on social network analysis: Trust relationship-based conflict detection and elimination | |
CA3014995C (en) | Searching for entities based on trust score and geography | |
US10127618B2 (en) | Determining connectivity within a community | |
US9679254B1 (en) | Extrapolating trends in trust scores | |
Zhao et al. | A machine learning based trust evaluation framework for online social networks | |
CN108664480B (en) | Multi-data-source user information integration method and device | |
WO2019173554A1 (en) | Blockchain transaction safety | |
CN103368917B (en) | A kind of risk control method and system of network virtual user | |
CN111274495B (en) | Data processing method, device, computer equipment and storage medium for user relationship strength | |
TW201643756A (en) | Calculating a trust score | |
WO2013097026A1 (en) | Systems and methods for visualizing social graphs | |
Weng et al. | An entropy-based approach to protecting rating systems from unfair testimonies | |
TW201530468A (en) | Obtaining social relationship type of network subjects | |
US20180033077A1 (en) | Method and system for generation of parameters | |
Ye | The generalized Dice measures for multiple attribute decision making under simplified neutrosophic environments | |
CN107622197A (en) | Device identification method and device, weighing computation method and device for equipment identification | |
Tommasel et al. | I want to break free! recommending friends from outside the echo chamber | |
Kaplan et al. | Trust estimation and fusion of uncertain information by exploiting consistency | |
CN103701939B (en) | Method for interchanging data | |
AlRubaian et al. | CredFinder: A real-time tweets credibility assessing system | |
Peng et al. | TCDABCF: A Trust‐Based Community Detection Using Artificial Bee Colony by Feature Fusion | |
Huang et al. | Using reputation measurement to defend mobile social networks against malicious feedback ratings | |
Oosthoek et al. | Quantifying dark web shops’ illicit revenue | |
Shreya et al. | Identification of Fake accounts in social media using machine learning | |
Carullo et al. | Friendship recommendations in online social networks |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant | ||
TR01 | Transfer of patent right | ||
TR01 | Transfer of patent right |
Effective date of registration: 20190912 Address after: 510000 No. 1101, No. 246, 248, 250, Wushan Road, Guangzhou City, Guangdong Province, self-compiled 08 locations Patentee after: Guangdong three hop Technology Investment Co., Ltd. Address before: 226019 Jiangsu Province, Nantong City Chongchuan District sik Road No. 9 Patentee before: Nantong University |