CN101707626A - Domain-based recommendation trust integration method in peer-to-peer network - Google Patents
Domain-based recommendation trust integration method in peer-to-peer network Download PDFInfo
- Publication number
- CN101707626A CN101707626A CN200910185426A CN200910185426A CN101707626A CN 101707626 A CN101707626 A CN 101707626A CN 200910185426 A CN200910185426 A CN 200910185426A CN 200910185426 A CN200910185426 A CN 200910185426A CN 101707626 A CN101707626 A CN 101707626A
- Authority
- CN
- China
- Prior art keywords
- trust
- territory
- recommendation
- node
- prestige
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 40
- 230000010354 integration Effects 0.000 title claims abstract description 16
- 230000003993 interaction Effects 0.000 claims abstract description 8
- 238000011156 evaluation Methods 0.000 claims description 10
- 238000013459 approach Methods 0.000 claims description 7
- 238000004364 calculation method Methods 0.000 claims description 7
- 238000000151 deposition Methods 0.000 claims description 3
- 230000005540 biological transmission Effects 0.000 description 4
- 230000000977 initiatory effect Effects 0.000 description 4
- 230000003542 behavioural effect Effects 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 230000002452 interceptive effect Effects 0.000 description 2
- 239000000203 mixture Substances 0.000 description 2
- 238000011160 research Methods 0.000 description 2
- 230000002159 abnormal effect Effects 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000009472 formulation Methods 0.000 description 1
- 230000002265 prevention Effects 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 230000001902 propagating effect Effects 0.000 description 1
- 230000000630 rising effect Effects 0.000 description 1
- 230000002123 temporal effect Effects 0.000 description 1
- 238000012546 transfer Methods 0.000 description 1
Images
Landscapes
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention provides a domain-based recommendation trust integration method in a peer-to-peer network, which is a strategic method mainly used for solving recommendation trust integration in a trust model based on a P2P network in an open network environment so as to solve the trust problem among P2P nodes. The method can improve the credibility and effectiveness of recommendation trust and reduce risk in P2P network interaction. The method provides a more reasonable and effective integration algorithm, subdivides the recommendation trust values collected and obtained in the network according to the differences of service types and interaction time, and leads the result of the trust integration to be more reasonable and effective by analyzing the value relationship and variation trend as well as different importance level among the recommend trust values.
Description
Technical field
The present invention is a kind of in open network environment, be mainly used in solution based on the integrated tactic method of the recommendation trust in the trust model of P2P network, this method provides a kind of more rational and effective integrated algorithm, belongs to the interleaving techniques application of equity calculating and information security.
Background technology
P2P (Peer-to-Peer) network is also referred to as peer-to-peer network, and it is not a completely new concept, and in fact it is the reflection of Internet substantive characteristics, and it is a kind of pattern of information resources transmission exchange, also is a kind of thought of network configuration.Peer-to-peer network directly gets up people contact, and it is directly mutual to allow people pass through the Internet.The P2P technology greatly improves the utilance to information, bandwidth and computational resource among the Internet.The P2P peer-to-peer network has been broken traditional Client, and the unit of each node all is identical in the peer-to-peer network, and each node had both served as server, for other nodes provide service, also serves as client computer simultaneously, enjoys the service that other nodes provide.
The P2P network is based on node and is ready shared resource this basic assumption, but actual situation is the independent behaviour of node makes node capacity show very big isomerism, studies show that in the Gnutella network that 70% node is the free-rider node, these nodes are only consumed the resource of other node contribution, and do not share the resource of oneself.In addition, exist a large amount of insecure service quality and fraud in the P2P network at present.Be applied as example with numerous file-sharings, 25% file is forge document (faked files).The uncooperative property that this independence of node causes has had a strong impact on the quality of P2P service.
What trust reflection is the overall merit of a user to another user behavior and ability, and the research to trust problem in the P2P network mainly is to set up reliable trust management model to solve in system at present.And in a large-scale P2P network, node will the acquisition system in the information of other all nodes are unusual difficulties.When needs and a strange node are mutual, must understand its prestige earlier, determine whether carrying out alternately with this.And can obtain indirectly by the recommendation of other node about the information of this strange node.During promptly in the face of a strange node, the transmission by recommendation trust and integrated just can be in the hope of the prestige of this node.
In fact, in the P2P network, the probability that carries out repeated interaction between two peer node is very low, and under a lot of situations, people are merely able to those strange fully nodes mutual, to obtain the resource that they need.In this case, by recommendation trust, can allow a node and another node when directly not mutual, also can make evaluation to the trusting degree of another node, try to achieve a kind of prestige reference of destination node, people are avoided risk, refusal is mutual with those bad nodes.But the recommendation trust value might not be exactly the true reflection of nominator to presentee's node confidence, and some recommended nodes may be exaggerated to the degree of belief of recommended node for different purposes and belittle.How to design a kind of rationally and effective method makes the integrated prestige of trying to achieve of recommendation trust more accurately with credible, also be one of focus of present network safety filed research.
Summary of the invention
Technical problem: the purpose of this invention is to provide in a kind of peer-to-peer network recommendation trust integration method based on the territory, solve the trust problem between the P2P node, the method that the application of the invention proposes can improve the credible and significant degree of recommendation trust, reduces the risk in the P2P network interaction.
Technical scheme: method of the present invention is a kind of method of tactic, the recommendation trust that third party's node is provided carries out the comparison between temporal classification and the trust value, and by suitable numerical value processing, make recommendation trust have very high confidence level, be enough to become the effective voucher of two strange nodes when mutual.
At first provide relevant notion and definition:
The territory: the node in the P2P network is clustered into the territory according to factors such as physical distance proximity or interest.All have Centroid in each territory, deposit various information and as the external mutual interface of member in the territory.
Centroid has been deposited the interactive information of whole territories interior nodes, and interactive information is represented with hexa-atomic group: (service requester; The ISP; Territory, supplier place; COS; The mutual time; The mutual evaluation).
Service vector: having service miscellaneous in P2P network mutual, these COS are collected and arrange in a certain order, is exactly the service vector.Representation is: S=(s
1, s
2..., s
n).S wherein
1To s
nBe whole COS in the P2P network.The element that subscript is adjacent all has certain contact, and such as " movie download ", " novel download ", " scientific and technical literature download ", these services will be adjacent in regular turn or be separated by very near.The service vector uses extensively in the method.Its element s
iValue can be 0 or 1, whether expression service exists, also can be the positive integer in the certain limit, the grade that expression is served.
Each territory all has a service vector, shows the grade of the service that the node in this territory can provide, and generations such as quantity that this grade provides according to the service of this domain node and cycle and are trusted feedback and had nothing to do.
Time threshold: a given time range T, recommendation trust is divided into two parts, give different weights respectively and participate in trusting integrated.
Current prestige and historical prestige: the prestige by the integrated acquisition of recommendation trust in the time threshold is current prestige; Prestige by the integrated acquisition of recommendation trust outside the time threshold is historical prestige.
Recommendation trust integration method based on the territory in the peer-to-peer network of the present invention will be collected the recommendation trust value that obtains in the network, according to COS, the difference of mutual time is segmented, and by analyzing the numerical relation and variation tendency and different significance levels between the recommendation trust value, make and trust integrated result more rationally effectively, concrete steps are as follows:
Step 1). node a sends to the Centroid of territory, place A with query requests, and Centroid is received query requests, and solicited message is converted to the query requests of territory to the territory, and this request is sent to other territory in the network,
Step 2). all receive the territory of query requests, check the database in this territory, if preserve with b node directly mutual after evaluation result, then these data are taken out, as trusting evidence,
Step 3). to depositing the arbitrarily-shaped domain I of b node interaction results in the step 2, gather the whole trust evidences in the territory, and, it is carried out a subseries according to the COS of trusting in the evidence,
Step 4). to sorted trust evidence cluster once, the time threshold according to setting carries out secondary classification to it, is divided into nearest trust evidence and in the past trust evidence,
Step 5). according to corresponding formulas, nearest trust evidence and in the past trust evidence are carried out integrated respectively, it is calculated that out current prestige by nearest credentials, and it is calculated that out historical prestige by credentials in the past,
Step 6). by the value of current prestige CR and historical prestige HR, by formula
Calculate integrated weights λ, weighted average calculation goes out initial recommendation trusts,
Step 7). territory I trusts the initial recommendation of all COS and carries out integratedly, establishes k and be the specified services in the query requests, and i is that initial recommendation is trusted corresponding COS, and n is the quantity of whole COS; According to formula
Calculate the service weights, weighted average calculation goes out recommendation trust in the territory,
Step 8). all calculate the territory of recommendation trust in the territory, and the result is passed to territory A, by hamming approach degree formula, calculate the territory similarity in each territory by territory A, and weighted average calculation goes out final recommendation trust, and the result is passed to node a,
Step 9). node a has obtained the recommendation trust integration value about node b, and overall process finishes.
In the computational process of historical prestige,
HR＝BSD*V+δ*BRD(HR∈[0，1])
V wherein is the historical weighted average that comprises the time attenuation function of trusting evidence, and BSD is a behavior stability index, and BRD is a behavior fluctuation degree index,
By setting up parameter δ, the oscillating-type node is distinguished from the GENERAL TYPE node, and by behavior trend BT and parameter θ, the value of δ has been done effectively defined; Be specially: during BT＞θ, δ=1; δ=0 during 1/ θ＜BT＜θ; δ=-1 during BT＜1/ θ, θ is the threshold value of the validity of behavior trend, θ＞1.
Beneficial effect: the present invention proposes a kind of integrated approach of recommendation trust value of combined type, different with trust integrated approach in the past, the present invention is segmented the recommendation trust that collection obtains, and divides afterwards earlier and closes, and trusts integrated validity thereby improved.Do below and specify:
At first, the integration problem of recommendation trust in the P2P network in the past, all be the form of node to be carried out with node, in large-scale P2P network, huge number of nodes is arranged, caused the transfer mode of this recommendation trust can only allow the initiation node of inquiry seek the opinion of the suggestion of the fraction node that closes on, make that integrated result's the confidence level of recommendation trust is not high, if the prolongation recommendation paths then can increase the load of network exponentially.In the method, with the integrated level of bringing up to the territory of recommendation trust, various trust data all are to be that unit transmits in network with the territory, have increased the range of propagating, and have also reduced the load of network.
Secondly, this method is classified to trusting evidence, divide for three levels, and refinement layer by layer.The integrated level of the integrated and final recommendation trust of recommendation trust in the territory, we have set up corresponding weights, make each group data can both obtain its due value.Wherein, in final recommendation trust integrated, adopted similarity degree between the territory as the formulation standard of weights, can recommendation trust be much accounted of more in the high territory that the territory provided of the similarity degree in territory so that those are initiated with inquiry.And in the territory in recommendation trust integrated, serve as to distinguish to have set up corresponding weights with the degree of closing on of COS, making that trusting integrated result is based on the trust of the special services of being inquired about, the trust of other COS is the pattern of reference.Can farthest embody the trust situation of target entity in special services.
And in the pattern that initial recommendation is trusted; this method will be trusted evidence at first in time and be distinguished; with time domain threshold values T is that boundary is divided into two recommendation trust; be divided into current recommendation trust and historical recommendation trust. because new more recommendation trust can embody the destination node behavioural characteristic of now more; therefore such classification can not be subjected to the influence of passing behavior and clear and observe the quality of the nearest behavior of destination node intuitively from current credit value. and historical recommendation trust is just as a people's archives; having write down the behavior of destination node in passing a period of time. this method is not simple these are trusted evidences carry out Mathematical treatment; but make every effort to from these data, to find the behavior rule of target entity in reciprocal process in the past; the stability of subordinate act; thereby many-sides such as the vibration situation of fluctuation degree and good and the bad behavior have been carried out the calculating and the analysis of science. can make objective correct evaluation to the historical behavior of destination node. particularly; in historical recommendation trust integrated; if the recommendation trust value reduces in time gradually; then integrated result can be subjected to the punishment (situation of BT＜1/ θ) on the numerical value; and the node that trust value is raise in time; integrated result can receive awards (situation of BT＞θ). and fluctuated for those recommendation trust values; Bo Dong entity up and down; their rising or downward trend are also not obvious; therefore can be not simple they are punished or reward; but integrated result can be compressed on one than (situation of 1/ θ＜BT＜θ) on the low slightly degree of the mean value of recommendation trust. after trying to achieve current prestige and historical prestige; this method uses weighted factor λ both to be carried out further integrated. and the setting of weighted factor λ; if considered current prestige and the different significance level of historical prestige in recommendation trust fully. current prestige is low; if instead then we can suspect the behavior situation that destination node is current to a large extent. current prestige height; but historical prestige is low; we also can't trust the behavior of destination node fully. therefore; our the value good treatment by λ this situation; make when current prestige is low; initial recommendation is trusted integration value convergence and current prestige; when current prestige was high, initial recommendation was trusted integration value and is leveled off to historical prestige.
This method is in the trust integrating process that with the territory is base unit, do not lose the independence and the independence of node itself yet, the territory at the initiation node place of recommendation trust inquiry, also be as a common territory, the participation of territory equality with other is trusted integrated, the weights that obtained when only being integrated are higher, and do not have other any privileges.
This method has been carried out many-sided comprehensive consideration to recommendation trust, has carried out rationally and concrete division trusting evidence, effectively raises the credible and significant degree of recommendation trust value.
Description of drawings
Fig. 1 is the reference architecture schematic diagram.The expression the inventive method based on the trust network structure.
Fig. 2 is a schematic flow sheet.The flow process signal of expression the inventive method.
Embodiment:
One. architecture
Figure one has provided and has used this method to carry out the integrated composition structure of recommendation trust.Different with general trust integrated approach, this method is not used the transmission carrier of node as recommendation trust, but with the integrated level that has risen between the territory of recommendation trust, is transmission and the integrating process that unit finishes recommendation trust with the territory.And by the similarity degree between territory and the territory and trust evidence based on different types of transaction, the corresponding weights that the recommendation trust value is divided and set up.Also considered simultaneously to trust the quality of evidence itself, it is high more that near more trust evidence of time is considered to quality.The integrating process of recommendation trust is divided into three parts in order: trust, initial recommendation is trusted by recommendation trust in the territory of the integrated acquisition of transaction weights, final recommendation trust by the integrated acquisition of trust between the territory based on the initial recommendation that different types of transaction are divided.
Below we provide the explanation of several concrete parts:
1. initial recommendation is trusted
It is after having collected whole trust evidences of territory interior nodes about the same service of destination node, it to be carried out the result of the integrated acquisition of science that initial recommendation is trusted.Its algorithm flow mainly is divided into three parts: current prestige is calculated, and historical prestige is calculated and both synthetic.
Current prestige (Current Reputation) is calculated: divided by time threshold, apart from time of current time recommendation trust value less than T, be current recommendation trust, their integrated result is current prestige.
The characteristics of current recommendation trust are that number is less, but can reflect the nearest behavioural characteristic of destination node, reference value height.
We use symbol V
i(T
i) represent that i third party's node is V to destination node at the trust value of time t, wherein the numerical value of t is that current time and this recommendation trust are given the difference of commenting the moment.The computing formula that provides current prestige is as follows:
Wherein n is the sum that falls into the recommendation trust of watch window.T is the time threshold values of watch window.D (t) is the time attenuation function, and formula is:
Historical prestige (Historical Reputation) is calculated: because the existence of time threshold T makes most recommendation trust all be put under historical prestige calculating, so algorithm is comparatively complicated with current prestige relatively.At first provide as giving a definition:
Behavior stability (Behavior Steady Degree): the behavior stability has been measured destination node whole behavior degree of stability in history in the past.Its value is big more, and the interbehavior of expression destination node is stable more.Its value uses the hamming approach degree to measure, and formula is as follows:
Behavior fluctuation degree (Behavior Rolling Degree): behavior fluctuation degree has reflected the average degree of fluctuation of historical recommendation trust value.Its numeric representation the unstable degree of destination node behavior, be worth greatly more, behavior is unstable more.Formula is as follows:
Behavior trend (Behavior Trend): behavior trend has reflected the trend variation that the destination node behavior improves or degenerates.If the evaluation of estimate of one time period of back is p greater than the total degree of previous time period evaluation of estimate, back one time period evaluation of estimate is n less than the total degree of previous time period evaluation of estimate.Then behavior trend formula is as follows:
BT＝p/n
Utilize above definition, the computing formula that then can obtain historical prestige is as follows:
HR＝BSD*V+δ*BRD(HR∈[0，1])
V wherein is the historical weighted average that comprises the time attenuation function of trusting evidence, and computing formula is with current prestige formula.The value of δ is relevant with BT, during BT＞θ, and δ=1; δ=0 during 1/ θ＜BT＜θ; δ=-1 during BT＜1/ θ.θ (θ＞1) is the threshold value of the validity of behavior trend, rewards in order to abnormal prestige that prevention oscillating-type node is obtained in the method.When HR＞1, perhaps HR＜0 o'clock, HR gets 1 or 0.
Current prestige and historical prestige integrated: after having obtained these two numerical value of current prestige and historical prestige, with they both in addition integrated, the initial recommendation that obtains k kind type of transaction is trusted formula and is:
R
k″＝λ*CR+(1-λ)*HR
λ wherein is integrated weights.Its value is:
2. recommendation trust in the territory
The calculating of service weights: a kind of COS that each initial recommendation trust is all corresponding, and the query requests of recommendation trust also corresponding a kind of specific service, be that the query requests node is prepared to carry out mutual service with destination node. the service weights have reacted the different COS and the weights ratio of the specified services in the query requests, because COS and service vector
In element corresponding one by one, the computing formula of service weights is:
K wherein is the specified services in the query requests, and i is that initial recommendation is trusted corresponding COS, and n is the quantity of whole COS.The formula that provides recommendation trust in the territory of territory j thus is as follows:
3. final recommendation trust
The territory calculation of similarity degree: each territory all has a service vector, shows the grade of the service that the node in this territory can provide, service vector that will inquiry initiation territory
Service vector with certain territory k
Do the calculating of approach degree, the result is the territory similarity, and formula is:
By recommendation trust in territory similarity and the territory, it is as follows to get final recommendation trust formula:
Two. algorithm flow
1., classify according to different COS taking out as the trust evidence about the interaction results of destination node in the territory in the territory of receiving query requests.
2. to the trust evidence of different COS, divide according to the time domain threshold values respectively, be divided into current prestige and historical prestige.
3. calculate current prestige and the historical prestige of recommending according to corresponding formulas.Again that both are integrated, obtain the initial recommendation trust value.
4. calculate the service weights that initial recommendation is trusted the corresponding COS of being inquired about of pairing COS.According to these service weights, the initial recommendation of different COS trusted to carry out secondary integrated, try to achieve recommendation trust in the territory.
5. calculate the territory similarity of the territory of recommendation trust in the territory that has this inquiry with respect to the initiation territory of inquiry.According to the territory similarity, to recommendation trust in the territory carry out three times integrated, try to achieve final recommendation trust.
For convenience of description, our supposition has following application example:
In the P2P network, node a belongs to territory A, he want with territory B in node b to carry out COS be s
iMutual, so initiate recommendation trust inquiry to node b.Execution in step is as follows:
1. node a sends to query requests the Centroid of territory, place A.Centroid is subjected to query requests, and solicited message is converted to the query requests of territory to the territory, and this request is sent to other territories in the network.
2. all are subjected to the territory of query requests, check the database in this territory, if preserve with b node directly mutual after evaluation result, then these data are taken out, as trusting evidence.
3. to depositing the arbitrarily-shaped domain 1 of b node interaction results in the step 2, gather the whole trust evidences in the territory, and, it carried out a subseries according to the COS of trusting in the evidence.
4. to sorted trust evidence cluster once, the time domain threshold values according to setting carries out secondary classification to it.Be divided into nearest trust evidence and trust evidence in the past.
5. according to corresponding formulas, nearest trust evidence and in the past trust evidence are carried out integrated respectively, it is calculated that out current prestige by nearest credentials, and it is calculated that out historical prestige by credentials in the past.
6. by the value of current prestige and historical prestige, calculate integrated weights λ,, calculate initial recommendation and trust according to corresponding formulas.
7. territory I trusts the initial recommendation of all COS and carries out integratedly, calculates recommendation trust in the territory according to respective formula.
8. all calculate the territory of recommendation trust in the territory, and the result is passed to territory A, by considering the territory similarity in each territory, according to respective formula, calculate final recommendation trust by territory A, and the result is passed to node a.
9. node a has obtained the recommendation trust integration value about node b, considers in conjunction with otherwise, judges whether with b mutual.
Claims (2)
- In the peer-to-peer network based on the recommendation trust integration method in territory, it is characterized in that this method will be in the network collects the recommendation trust value that obtains, according to COS, the difference of mutual time is segmented, and by analyzing the numerical relation and variation tendency and different significance levels between the recommendation trust value, make and trust integrated result more rationally effectively, concrete steps are as follows:Step 1). node a sends to the Centroid of territory, place A with query requests, and Centroid is received query requests, and solicited message is converted to the query requests of territory to the territory, and this request is sent to other territory in the network,Step 2). all receive the territory of query requests, check the database in this territory, if preserve with b node directly mutual after evaluation result, then these data are taken out, as trusting evidence,Step 3). to depositing the arbitrarily-shaped domain I of b node interaction results in the step 2, gather the whole trust evidences in the territory, and, it is carried out a subseries according to the COS of trusting in the evidence,Step 4). to sorted trust evidence cluster once, the time threshold according to setting carries out secondary classification to it, is divided into nearest trust evidence and in the past trust evidence,Step 5). according to corresponding formulas, nearest trust evidence and in the past trust evidence are carried out integrated respectively, it is calculated that out current prestige by nearest credentials, and it is calculated that out historical prestige by credentials in the past,Step 6). by the value of current prestige CR and historical prestige HR, by formulaCalculate integrated weights λ, weighted average calculation goes out initial recommendation trusts,
Step 7). territory I trusts the initial recommendation of all COS and carries out integratedly, establishes k and be the specified services in the query requests, and i is that initial recommendation is trusted corresponding COS, and n is the quantity of whole COS; According to formula Calculate the service weights, weighted average calculation goes out recommendation trust in the territory,Step 8). all calculate the territory of recommendation trust in the territory, and the result is passed to territory A, by hamming approach degree formula, calculate the territory similarity in each territory by territory A, and weighted average calculation goes out final recommendation trust, and the result is passed to node a,Step 9). node a has obtained the recommendation trust integration value about node b, and overall process finishes. - 2. based on the recommendation trust integration method in territory, it is characterized in that in the computational process of historical prestige in the peer-to-peer network according to claim 1,HR＝BSD*V+δ*BRD(HR∈[0，1])V wherein is the historical weighted average that comprises the time attenuation function of trusting evidence, and BSD is a behavior stability index, and BRD is a behavior fluctuation degree index,By setting up parameter δ, the oscillating-type node is distinguished from the GENERAL TYPE node, and by behavior trend BT and parameter θ, the value of δ has been done effectively defined; Be specially: during BT＞θ, δ=1; δ=0 during 1/ θ＜BT＜θ; δ=-1 during BT＜1/ θ, θ is the threshold value of the validity of behavior trend, θ＞1.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2009101854268A CN101707626B (en) | 2009-11-09 | 2009-11-09 | Domain-based recommendation trust integration method in peer-to-peer network |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2009101854268A CN101707626B (en) | 2009-11-09 | 2009-11-09 | Domain-based recommendation trust integration method in peer-to-peer network |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN101707626A true CN101707626A (en) | 2010-05-12 |
| CN101707626B CN101707626B (en) | 2012-04-11 |
Family
ID=42377816
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN2009101854268A Expired - Fee Related CN101707626B (en) | 2009-11-09 | 2009-11-09 | Domain-based recommendation trust integration method in peer-to-peer network |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN101707626B (en) |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101867608A (en) * | 2010-05-24 | 2010-10-20 | 中国科学院计算技术研究所 | Centralized block switching P2P (peer-to-peer) file sharing system and service method thereof |
| CN101895577A (en) * | 2010-07-06 | 2010-11-24 | 中国科学院计算技术研究所 | Distribution method of network sharing resources |
| CN102006305A (en) * | 2010-12-08 | 2011-04-06 | 上海交通大学 | Method for suppressing malicious behavior based on distributed recommendable reputation in peer-to-peer (P2P) network |
| CN102347917A (en) * | 2011-11-04 | 2012-02-08 | 西安电子科技大学 | Contact semantic grouping method for network message communication |
| CN102638795A (en) * | 2012-04-23 | 2012-08-15 | 浙江大学 | Trust evaluation method of distributed sensor network capable of resisting attacks |
| WO2012162873A1 (en) * | 2011-05-27 | 2012-12-06 | Nokia Corporation | Method and apparatus for role-based trust modeling and recommendation |
| CN104102648A (en) * | 2013-04-07 | 2014-10-15 | 腾讯科技（深圳）有限公司 | User behavior data based interest recommending method and device |
| CN104574148A (en) * | 2015-01-09 | 2015-04-29 | 西安融成科技有限公司 | Trust system establishment method for visual community |
Family Cites Families (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN100518086C (en) * | 2006-01-13 | 2009-07-22 | 华为技术有限公司 | Method of controlling coordinate network and its node |
| CN101404592A (en) * | 2008-11-14 | 2009-04-08 | 西安交通大学 | Total trust degree decision method based on history evidence window |
| CN101404572A (en) * | 2008-11-14 | 2009-04-08 | 西安交通大学 | Network node total trust degree estimation method based on feedback trust aggregation |
-
2009
- 2009-11-09 CN CN2009101854268A patent/CN101707626B/en not_active Expired - Fee Related
Cited By (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101867608B (en) * | 2010-05-24 | 2013-03-13 | 中国科学院计算技术研究所 | Centralized block switching P2P (peer-to-peer) file sharing system and service method thereof |
| CN101867608A (en) * | 2010-05-24 | 2010-10-20 | 中国科学院计算技术研究所 | Centralized block switching P2P (peer-to-peer) file sharing system and service method thereof |
| CN101895577A (en) * | 2010-07-06 | 2010-11-24 | 中国科学院计算技术研究所 | Distribution method of network sharing resources |
| CN101895577B (en) * | 2010-07-06 | 2013-02-13 | 中国科学院计算技术研究所 | Distribution method of network sharing resources |
| CN102006305A (en) * | 2010-12-08 | 2011-04-06 | 上海交通大学 | Method for suppressing malicious behavior based on distributed recommendable reputation in peer-to-peer (P2P) network |
| CN102006305B (en) * | 2010-12-08 | 2013-05-01 | 上海交通大学 | Method for suppressing malicious behavior based on distributed recommendable reputation in peer-to-peer (P2P) network |
| WO2012162873A1 (en) * | 2011-05-27 | 2012-12-06 | Nokia Corporation | Method and apparatus for role-based trust modeling and recommendation |
| CN102347917A (en) * | 2011-11-04 | 2012-02-08 | 西安电子科技大学 | Contact semantic grouping method for network message communication |
| CN102347917B (en) * | 2011-11-04 | 2014-04-16 | 西安电子科技大学 | Contact semantic grouping method for network message communication |
| CN102638795A (en) * | 2012-04-23 | 2012-08-15 | 浙江大学 | Trust evaluation method of distributed sensor network capable of resisting attacks |
| CN102638795B (en) * | 2012-04-23 | 2014-11-19 | 浙江大学 | Trust evaluation method of distributed sensor network capable of resisting attacks |
| CN104102648A (en) * | 2013-04-07 | 2014-10-15 | 腾讯科技（深圳）有限公司 | User behavior data based interest recommending method and device |
| CN104102648B (en) * | 2013-04-07 | 2017-12-01 | 腾讯科技（深圳）有限公司 | Interest based on user behavior data recommends method and device |
| CN104574148A (en) * | 2015-01-09 | 2015-04-29 | 西安融成科技有限公司 | Trust system establishment method for visual community |
Also Published As
| Publication number | Publication date |
|---|---|
| CN101707626B (en) | 2012-04-11 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN101707626B (en) | Domain-based recommendation trust integration method in peer-to-peer network | |
| Jiang et al. | Federated learning in smart city sensing: Challenges and opportunities | |
| Shen et al. | Blockchain-based incentives for secure and collaborative data sharing in multiple clouds | |
| Zhu et al. | Blockchain-empowered federated learning: Challenges, solutions, and future directions | |
| Recalde et al. | Structural energy poverty vulnerability and excess winter mortality in the European Union: Exploring the association between structural determinants and health | |
| Otoum et al. | A federated learning and blockchain-enabled sustainable energy trade at the edge: A framework for industry 4.0 | |
| Zhao et al. | Mobile edge computing, blockchain and reputation-based crowdsourcing IoT federated learning: A secure, decentralized and privacy-preserving system | |
| Heckathorn | Respondent-driven sampling II: deriving valid population estimates from chain-referral samples of hidden populations | |
| Xiao et al. | BayesTrust and VehicleRank: Constructing an implicit Web of trust in VANET | |
| Dong et al. | Social bridges in urban purchase behavior | |
| Jia et al. | A trust and reputation model considering overall peer consulting distribution | |
| Meng et al. | @ Trust: A trust model based on feedback-arbitration in structured P2P network | |
| Fabris et al. | Measuring fairness under unawareness of sensitive attributes: A quantification-based approach | |
| Wang et al. | A fast and secured vehicle-to-vehicle energy trading based on blockchain consensus in the internet of electric vehicles | |
| Wang et al. | Trustworthy and efficient crowdsensed data trading on sharding blockchain | |
| Din et al. | ShareTrust: Centralized trust management mechanism for trustworthy resource sharing in industrial Internet of Things | |
| Zhao et al. | A survey of sparse mobile crowdsensing: Developments and opportunities | |
| Tan et al. | Digital twin based remote resource sharing in internet of vehicles using consortium blockchain | |
| Chen et al. | A blockchain-based creditable and distributed incentive mechanism for participant mobile crowdsensing in edge computing | |
| Liu et al. | Understanding the decentralization of DPoS: Perspectives from data-driven analysis on EOSIO | |
| Li et al. | The study of trust vector based trust rating aggregation in service-oriented environments | |
| Wang et al. | Distributed game-theoretical route navigation for vehicular crowdsensing | |
| Guo | Trust-based service management of internet of things systems and its applications | |
| Gupta et al. | Growth of research literature in scientific specialities. A modelling perspective | |
| CN115640305A (en) | Fair and credible federal learning method based on block chain |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| EE01 | Entry into force of recordation of patent licensing contract |
Application publication date: 20100512 Assignee: Jiangsu Nanyou IOT Technology Park Ltd. Assignor: NANJING University OF POSTS AND TELECOMMUNICATIONS Contract record no.: 2016320000219 Denomination of invention: Domain-based recommendation trust integration method in peer-to-peer network Granted publication date: 20120411 License type: Common License Record date: 20161121 |
|
| LICC | Enforcement, change and cancellation of record of contracts on the licence for exploitation of a patent or utility model | ||
| EC01 | Cancellation of recordation of patent licensing contract |
Assignee: Jiangsu Nanyou IOT Technology Park Ltd. Assignor: NANJING University OF POSTS AND TELECOMMUNICATIONS Contract record no.: 2016320000219 Date of cancellation: 20180116 |
|
| EC01 | Cancellation of recordation of patent licensing contract | ||
| EE01 | Entry into force of recordation of patent licensing contract |
Application publication date: 20100512 Assignee: JIANGSU I-FRONT SCIENCE & TECHNOLOGY Co.,Ltd. Assignor: NANJING UNIVERSITY OF POSTS AND TELECOMMUNICATIONS Contract record no.: X2020980000370 Denomination of invention: Domain-based recommendation trust integration method in peer-to-peer network Granted publication date: 20120411 License type: Exclusive License Record date: 20200227 |
|
| EE01 | Entry into force of recordation of patent licensing contract | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20120411 Termination date: 20211109 |
|
| CF01 | Termination of patent right due to non-payment of annual fee |