CN104717244A - Multidimensional credibility management method based on distributed computation - Google Patents
Multidimensional credibility management method based on distributed computation Download PDFInfo
- Publication number
- CN104717244A CN104717244A CN201310681954.9A CN201310681954A CN104717244A CN 104717244 A CN104717244 A CN 104717244A CN 201310681954 A CN201310681954 A CN 201310681954A CN 104717244 A CN104717244 A CN 104717244A
- Authority
- CN
- China
- Prior art keywords
- information
- node
- reputation information
- section point
- reputation
- 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
Abstract
The invention relates to the field of computer networks and provides a multidimensional credibility management method based on distributed computation so that credibility data between users can be more accurate, reliable and practical. The method includes the steps that a multidimensional hierarchy model is established, and characteristic vectors in the model are set; a first node acquires trade information and performs assignment on the characteristic vectors in the multidimensional hierarchy model so that initial credibility information can be generated; a first node sends the initial credibility information to a second node in real time; the second node receives the initial credibility information sent by the first node in real time; the second node performs mean value computation on the initial credibility information so that multidimensional credibility information can be obtained. According to the method, due to the establishment of a multidimensional credibility management mechanism, the credibility data are sent and received between neighbor nodes in real time, and the neighbor nodes perform mean value computation on the credibility data so that the credibility data between the users can be more accurate, reliable and practical.
Description
Technical field
The present invention relates to computer network field, particularly relate to a kind of various dimensions credit management method based on Distributed Calculation.
Background technology
Prestige Management is the major issue in socio-economic activity.Concrete, a kind of Prestige Management mechanism is that the prestige that each individuality that may participate in certain economic activity carries out being correlated with is evaluated and safeguards.When related economic activity occurs, related personnel can with reference to the degrees of comparison of the other side in this Prestige Management mechanism, and auxiliary its makes a policy.Because this degrees of comparison produces important impact by the economic life of reality, therefore it must possess the feature of " reliability ", and what namely the degrees of comparison of a people should be correct reflects the credibility of this person in economic activity.
The requirement of " reliability ", makes degrees of comparison should meet the feature of " fairness ", " ageing " in the process of collection, evaluation.That is, the node of any related economic activity, does not possess chance and true amendment is run counter in the evaluation of condition to degrees of comparison; And the evaluation of degrees of comparison should carrying out along with socio-economic activity, correctly can be reflected in current time, the credibility of economic activity node.
Due to the requirement of " reliability " and " fairness ", traditional Prestige Management mechanism needs the third party (as government department) relied on outside economic activity to carry out centralized management.Centralized Prestige Management can play just prestige evaluation and the effect of supervision to a certain extent, but still has following several deficiency.
Evaluation resource consolidation difficulty: as the third party implementing centralized Prestige Management, its first problem first faced is the evaluation that utilize which type of resource and information to carry out prestige.Usually, such resource is diversified.How they being combined, and generate a unique degrees of comparison, is a very difficult problem.At present, common way first determines that safeguarded degrees of comparison carries out the actual environment applied, and then according to the correlation of different information resources and this actual environment, judges the desirable degree of its information.As in current bank's Prestige Management mechanism, loan is given back whether timely, be classified as the key factor affecting prestige.Thisly first determine environment, although there is reasonability in the way of according to circumstances determination information importance, also there is strong specific aim simultaneously.Very naturally, the degrees of comparison obtained like this, is only applicable to predetermined environment.And, be much information resource, determine the correlation of itself and applied environment, also there is general difficulty.
Information channel is narrow: due to the requirement of " fairness ", and the information resources judged as degrees of comparison must be obtained by the collection channel of holding qualification.And this point just determines, the channel obtaining this type of information resource can not be very extensive.Face current network size explosion, the situation that information explosion formula increases, narrow information channel, what meeting was potential causes the incomplete of information, or not in time.
System maintenance cost is huge: because reputation information is managed by centralized mode, this just proposes very high technical requirement to the third party managed.Face the information of flood tide, it is examined, judge, analyze even to store and all mean huge cost payout.And in fact, the Prestige Management realized by centralized management, is accepting prestige query aspects, also there is the limit of ability, therefore may cause delay during reputation query.
Summary of the invention
The invention provides a kind of various dimensions credit management method based on Distributed Calculation, thus make the reputation data between user more practical.
The present invention adopts following technical scheme:
Based on a various dimensions credit management method for Distributed Calculation, for the Prestige Management in Transaction Information, comprising:
Set up various dimensions Grade Model, the characteristic vector in setting model;
First node gathers Transaction Information, generates initial reputation information;
First node sends described initial reputation information in real time to Section Point;
The initial reputation information that Section Point real-time reception first node sends;
Section Point is integrated initial reputation information, draws integration reputation information;
Section Point, according to user's request, is resolved and is integrated reputation information, obtain various dimensions reputation information.
Preferably, described Transaction Information is the transaction data that both parties' dealing produces.
Preferably, after first node collects Transaction Information, according to Transaction Information with from the characteristic vector group that maintenance server obtains, initial multidimensional reputation vectors can be calculated, generates initial reputation information.
Preferably, first node and Section Point neighbor node each other.
Preferably, the initial reputation information of described Section Point to real-time reception carries out integration and comprises:
The mean value computation of the initial reputation information of explicit dimension, obtains and integrates reputation information;
The detection of implicit expression dimension and the mean value computation to the initial reputation information of implicit expression dimension after detection, obtain and integrate reputation information.
Preferably, the detection of described implicit expression dimension comprises: set up the set of initial reputation information characteristic vector, characteristic vector is obtained by principal component analysis processing method, and as the characteristic vector of degrees of comparison dimension, with the characteristic vector of degrees of comparison dimension for basis vector, calculate the coordinate of various dimensions information eigenvector under new basement feature vector as the initial reputation information of implicit expression dimension.
Preferably, described parsing is integrated reputation information and is referred to, described Section Point is resolved according to the vector of described first node reputation information, to obtain the multidimensional reputation information of described first node.
Preferably, after the mean value computation that described explicit dimension detects goes out result, Section Point is filed a request, and utilizes the dimension analyzed and obtain, calculates the confidence level of first node, and send to Section Point.
Preferably, after the mean value computation that described implicit expression dimension detects goes out result, Section Point is filed a request, and finds characteristic of correspondence vector and reputation vectors, calculate reputation information, and send to Section Point according to the request that Section Point proposes.
Preferably, to the described reputation information calculated, by the inner product computational methods between vector, draw various dimensions reputation information.
Relative with prior art, beneficial effect of the present invention is, by the various dimensions credit management method based on Distributed Calculation, sends in real time and receive reputation data between neighbor node, neighbor node carries out mean value computation to reputation data, makes the reputation data between user more practical.
Accompanying drawing explanation
Fig. 1 is the flow chart of a kind of various dimensions credit management method based on Distributed Calculation of the present invention;
Fig. 2 is the average calculation method of the explicit dimension detection of the embodiment of the present invention 1;
Fig. 3 is the method for the embodiment of the present invention 2 implicit expression dimension detection.
Embodiment
In order to make object of the present invention, technical scheme and advantage clearly understand, below in conjunction with drawings and Examples, the present invention is further elaborated.Should be appreciated that specific embodiment described herein only in order to explain the present invention, be not intended to limit the present invention.
Please refer to Fig. 1, a kind of various dimensions credit management method based on Distributed Calculation, for the Prestige Management in Transaction Information, comprising:
Set up various dimensions Grade Model, the characteristic vector in setting model;
First node gathers Transaction Information, generates initial reputation information;
First node sends described initial reputation information in real time to Section Point;
The initial reputation information that Section Point real-time reception first node sends;
Section Point is integrated initial reputation information, draws integration reputation information;
Section Point, according to user's request, is resolved and is integrated reputation information, obtain various dimensions reputation information.
Preferably, described Transaction Information is the transaction data that both parties' dealing produces.
Preferably, after first node collects Transaction Information, according to Transaction Information with from the characteristic vector group that maintenance server obtains, initial multidimensional reputation vectors can be calculated, generates initial reputation information.
Preferably, first node and Section Point neighbor node each other.
Preferably, the initial reputation information of described Section Point to real-time reception carries out integration and comprises:
The mean value computation of the initial reputation information of explicit dimension, obtains and integrates reputation information;
The detection of implicit expression dimension and the mean value computation to the initial reputation information of implicit expression dimension after detection, obtain and integrate reputation information.
Preferably, the detection of described implicit expression dimension comprises: set up the set of initial reputation information characteristic vector, characteristic vector is obtained by principal component analysis processing method, and as the characteristic vector of degrees of comparison dimension, with the characteristic vector of degrees of comparison dimension for basis vector, calculate the coordinate of various dimensions information eigenvector under new basement feature vector as the initial reputation information of implicit expression dimension.
Preferably, described parsing is integrated reputation information and is referred to, described Section Point is resolved according to the vector of described first node reputation information, to obtain the multidimensional reputation information of described first node.
Preferably, after the mean value computation that described explicit dimension detects goes out result, Section Point is filed a request, and utilizes the dimension analyzed and obtain, calculates the confidence level of first node, and send to Section Point.
Preferably, after the mean value computation that described implicit expression dimension detects goes out result, Section Point is filed a request, and finds characteristic of correspondence vector and reputation vectors, calculate reputation information, and send to Section Point according to the request that Section Point proposes.
Preferably, to the described reputation information calculated, by the inner product computational methods between vector, draw various dimensions reputation information.
Relative with prior art, beneficial effect of the present invention is, by the various dimensions credit management method based on Distributed Calculation, sends in real time and receive reputation data between neighbor node, neighbor node carries out mean value computation to reputation data, makes the reputation data between user more practical.
Embodiment 1:
Please refer to Fig. 1 and Fig. 2, the embodiment of the present invention 1 provide a kind of based in the various dimensions credit management method of Distributed Calculation for the reputation information processing method that explicit dimension detects.
Be example with shopping at network, be illustrated in prestige dimension and can show when determining, the execution mode of the scheme in this patent.
According to current shopping at network common practice, the information that buyer comparatively pays close attention to has " product quality ", " logistics speed ", " product quality " and " after-sale service " several aspect.For setting forth facility, we get these three aspects as an example.And seller compares it is of concern that " the payment speed " of buyer, aspects such as " satisfactions ".
Step S1: set up various dimensions Grade Model, the characteristic vector in setting model.
In a model, we are by the degrees of comparison of each user, carry out representing (as a, b, c, d, e) with 5 dimensional feature vectors.Wherein, whether reaction product that businessman sells is met it by digital a describes, and whether has the problem of quality aspect; Whether projection seller can deliver by numeral b in time, and whether buyer can obtain product within reasonable time; Whether numeral c will react seller and process proper in service side face after sale; Numeral d reacts user as buyer, whether confirms transaction in time, completes payment; Numeral e reacts user as buyer, whether has irrational after-sale service requirement after completing transaction.
Step S2: first node gathers Transaction Information, and carries out assignment to the characteristic vector in described various dimensions Grade Model, generates initial reputation information.
In distributed reputation management method, each buyer or seller can be nodes.By adding up its transaction account information (as Taobao's information, Alipay information), when having concluded the business each time, information will be carried out as follows:
As the buyer of transaction, if transaction completes smoothly and confirms, we can assert that the goods of the seller conforms to quality requirements, and therefore remember a=1, otherwise, note a=0; Calculate transaction to start, to the time difference terminated, and to make b equal this time difference, with the speed of this combined reaction seller delivery and logistics; If after a transaction, the buyer proposes to complain or similar behavior, and we think that the seller may have problems in service side face after sale, make c be 1, otherwise c are 0.So, for once concluding the business, can obtain (a, b, c, *, *) such information about seller's prestige, wherein, rear two dimensions are due to inapplicable, thus default.
Similar, as the seller of transaction, if we calculate the buyer by confirming transaction, to the time difference of actual payment, make the d time difference for this reason; If the seller proposes to complain or similar behavior after a transaction, we think the requirement of the buyer in product or service may there is irrational composition, therefore remember that e is 1, otherwise note e is 0.Like this, by concluding the business each time, we will obtain (*, *, *, d, e) such information for buyer's prestige.
Step S3: first node sends described initial reputation information in real time to Section Point.
Be in each node in network topology structure, carry out the transmission of nodal information in real time and collect.A node, held about other possessory reputation informations, by specific network topology structure, pass to its neighbor node in a network.
Step S4: the initial reputation information that Section Point real-time reception first node sends.
Each node constantly receives the information of being sent by its neighbor node.
Step S5: Section Point carries out mean value computation to initial reputation information, draws various dimensions reputation information;
After a node obtains information from other nodes, need to integrate it.Specifically, when a node receives many reputation informations about some nodes, node needs to integrate these information.Use the method for mean value calculation, multiple information is merged.Namely for each component in got information eigenvector, new value is obtained by the method for calculating mean value.Node A obtains (collect from other nodes, or oneself and B conclude the business after calculate) about three prestige records of Node B: (1,3,1, *, *), (0,5,0, *, *), (1,1,0, *, *).After carrying out mean value computation, obtain a reputation information about Node B, (2/3,7/3,1/3).For certain value, do not add mean value computation.Afterwards, the information about Node B after integration only sends by node A, and three characteristic vectors before directly abandon.
At any time, what each node stored is from the information that other nodes obtain in last round of information exchanging process, and the information that in the transaction just occurred, its collects itself.This point ensure that as any node, and its storage information content is maintained at the rank close to linear complexity, thus can not cause the pressure of memory space.
When after the reputation information that node A obtains about Node B, needing the situation according to practical application, to representing that 5 dimensional vectors (a, b, c, d, e) of reputation information are resolved, drawing various dimensions degrees of comparison information.
When node A needs to obtain the information about Node B, a kind of may be that node A is at the last round of reputation information just obtained about Node B.In this case, A can aspect carry out information inquiry.Another kind may be, A is at the last round of reputation information that there is no Node B.In this case, A needs to wait for and severally takes turns the time.When credit mechanism topology of networks meets social networks feature, A (according to the conclusion of correlative study to social networks diameter, about usually only needing 10 to take turns) will can obtain the information about Node B within the antipode short time.Due to the method that this patent proposes, in commission based on distributed computer program, each time of taking turns is very of short duration.Therefore, in order to wait for the lower information about Node B, node A does not need to pay a lot of time.And due in economic activity, trade network has the feature of community usually, between the user of i.e. mutual dealing, formed and relatively connect sub-network structure closely, major part transaction to degrees of comparison produce need time, the information of interdependent node just can obtain within very likely taking turns 1 to 2.
Deeper mode is, Section Point is resolved result, and Section Point is filed a request, and utilizes the dimension obtained, calculates the confidence level of first node, and send to Section Point.
Node is as the degrees of comparison of seller, in fact very strong relevance is had with it as some information during buyer, therefore the request that will propose according to node A of system, as asked seller's reputation information of B, what utilize acquisition has informative several dimension, calculating about the current confidence level as seller of Node B, as used a real number be between 0 to 1, and being reported to node A.Node A according to the size of this number, can directly and effectively understand the trusted degree of Node B as seller.
Embodiment 2
Please refer to Fig. 1 and Fig. 3, in this example, we are still example with shopping at network, and the dimension set forth when degrees of comparison is the situation that implicit expression dimension detects.
According to current shopping at network common practice, the information that buyer comparatively pays close attention to has " product quality ", " logistics speed ", " product quality " and " after-sale service " several aspect.For setting forth facility, we get these three aspects as an example.And seller compares it is of concern that " the payment speed " of buyer, aspects such as " satisfactions ".
Step S1: set up various dimensions Grade Model, the characteristic vector in setting model.
In a model, we are by the degrees of comparison of each user, carry out representing (a, b, c, d, e) with 5 dimensional vectors.Wherein, whether reaction product that businessman sells is met it by digital a describes, and whether has the problem of quality aspect; Whether projection seller can deliver by numeral b in time, and whether buyer can obtain product within reasonable time; Whether numeral c will react seller and process proper in service side face after sale; Numeral d reacts user as buyer, whether confirms transaction in time, completes payment; Numeral e reacts user as buyer, whether has irrational after-sale service requirement after completing transaction.
Step S2: first node gathers Transaction Information, and carries out assignment to the characteristic vector in described various dimensions Grade Model, generates initial reputation information.
In distributed reputation administrative mechanism, each buyer or seller can be nodes.By adding up its trading account (as Taobao's information, Alipay information), when having concluded the business each time, the mechanism that this patent formula proposes, will carry out information as follows.
As the buyer of transaction, if transaction completes smoothly and confirms, we can assert that the goods of the seller conforms to quality requirements, and therefore remember a=1.Otherwise, note a=0; Calculate transaction to start, to the time difference terminated, and to make b equal this time difference, with the speed of this combined reaction seller delivery and logistics; If after a transaction, the buyer proposes to complain or similar behavior, and we think that the seller may have problems in service side face after sale, make c be 1, otherwise c are 0.So, for once concluding the business, can obtain (a, b, c, *, *) such information about seller's prestige, wherein, rear two dimensions are due to inapplicable, thus default.
Similar, as the seller of transaction, if we calculate the buyer by confirming transaction, to the time difference of actual payment, make the d time difference for this reason; If the seller proposes to complain or similar behavior after a transaction, we think the requirement of the buyer in product or service may there is irrational composition, therefore remember that e is 1, otherwise note e is 0.Like this, by concluding the business each time, we will obtain (*, *, *, d, e) such information for buyer's prestige.
Step S3: first node sends described initial reputation information in real time to Section Point.
Be in each node in degrees of comparison mechanism, enter sending out and collecting of passerby's information in real time.A node, held about other possessory reputation informations, by specific network topology structure, pass to its neighbor node in a network.
Step S4: the initial reputation information that Section Point real-time reception first node sends.
Each node constantly collects the information of being sent by its neighbor node.
Step S5: Section Point carries out mean value computation to initial reputation information, draws various dimensions reputation information;
Set up an initial reputation information set.In the example of electronic business transaction, i.e. a set be made up of the vector of 5 dimensions.Vector in this set has openness feature.I.e. most information, has several component to be disappearance.According to the theory of characteristic vector detection, in most of activity, effective information dimension is well below obtaining prestige dimension.In this example, namely can suppose that actual effective degrees of comparison dimension number will lower than 5.
In order to obtain actual effective information dimension, to existing information vector set, the processing method of principal component analysis can be used.These class methods are widely used in the field of sparse matrix dimensionality reduction.Its principle is that to form a columns be the matrix of 5 by existing 5 dimension information vectors, by the covariance matrix to this matrix, carries out analysis of spectrum calculating, obtains the characteristic vector of corresponding notable feature value.Direction representated by these characteristic vectors, can be used as the characteristic vector of degrees of comparison dimension.With these characteristic vectors, number is less than 5, is base vector, and we can calculate existing 5 dimension information vectors at new subbasal coordinate.And the final degrees of comparison vector used using this coordinate as system, such as 3 dimensions.
When carrying out information integration, what node was held has two classes vectors, and a class is the reputation vectors of 3 dimensions obtained from other nodes, also has a class to be the information vectors of 5 dimensions that this node is collected.In order to carry out information integration, first this node is by the dimensional characteristics vector according to reputation model, i.e. basis vector, calculates its 5 dimension information vectors of collecting at subbasal coordinate, and obtains 3 dimension reputation vectors of its correspondence.
Afterwards, the 3 dimension reputation vectors about same node that this node will be held it, integrate in the mode of mean value calculation, and the reputation vectors that will newly obtain, send to the neighbor node in network.
In information extraction, node A proposes the information (i.e. the information of a reaction in information vector) wanting to understand Node B " product quality ", system is according to the characteristic vector of product quality in degrees of comparison system (i.e. the first dimensional vector of 5 dimension information vectors, projection/feature in 3 dimension prestige dimensions), and 3 dimension reputation vectors of Node B, calculate a confidence level parameter, the computational methods of confidence level parameter, the inner product between vector can be used to calculate.And this parameter is returned to node A, in this, as multidimensional degrees of comparison information.
Relative with prior art, beneficial effect of the present invention is, by the various dimensions credit management method based on Distributed Calculation, sends in real time and receive reputation data between neighbor node, neighbor node carries out mean value computation to reputation data, makes the reputation data between user more practical.
The foregoing is only preferred embodiment of the present invention, not in order to limit the present invention, all any amendments done within the spirit and principles in the present invention, equivalent replacement and improvement etc., all should be included within protection scope of the present invention.
Claims (10)
1., based on a various dimensions credit management method for Distributed Calculation, for the Prestige Management in Transaction Information, it is characterized in that, comprising:
Set up various dimensions Grade Model, the characteristic vector in setting model;
First node gathers Transaction Information, generates initial reputation information;
First node sends described initial reputation information in real time to Section Point;
The initial reputation information that Section Point real-time reception first node sends;
Section Point is integrated initial reputation information, draws integration reputation information;
Section Point, according to user's request, is resolved and is integrated reputation information, obtain various dimensions reputation information.
2. method according to claim 1, is characterized in that, described Transaction Information is the transaction data that both parties' dealing produces.
3. method according to claim 1, is characterized in that, after first node collects Transaction Information, according to Transaction Information with from the characteristic vector group that maintenance server obtains, can calculate initial multidimensional reputation vectors, generate initial reputation information.
4. method according to claim 1, is characterized in that, first node and Section Point neighbor node each other.
5. method according to claim 1, is characterized in that, the initial reputation information of described Section Point to real-time reception carries out integration and comprise:
The mean value computation of the initial reputation information of explicit dimension, obtains and integrates reputation information;
The detection of implicit expression dimension and the mean value computation to the initial reputation information of implicit expression dimension after detection, obtain and integrate reputation information.
6. method according to claim 4, it is characterized in that, the detection of described implicit expression dimension comprises: set up the set of initial reputation information characteristic vector, characteristic vector is obtained by principal component analysis processing method, and as the characteristic vector of degrees of comparison dimension, with the characteristic vector of degrees of comparison dimension for basis vector, calculate the coordinate of various dimensions information eigenvector under new basement feature vector as the initial reputation information of implicit expression dimension.
7. method according to claim 1, is characterized in that, described parsing is integrated reputation information and referred to, described Section Point is resolved according to the vector of described first node reputation information, to obtain the multidimensional reputation information of described first node.
8. method according to claim 1 or 5, is characterized in that, after the mean value computation of described explicit dimension detection goes out result, Section Point is filed a request, and utilizes the dimension analyzed and obtain, calculates the confidence level of first node, and send to Section Point.
9. method according to claim 5, is characterized in that, after the mean value computation that described implicit expression dimension detects goes out result, Section Point is filed a request, find characteristic of correspondence vector and reputation vectors according to the request that Section Point proposes, calculate reputation information, and send to Section Point.
10. method according to claim 6, is characterized in that, to the described reputation information calculated, by the inner product computational methods between vector, draws various dimensions reputation information.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310681954.9A CN104717244B (en) | 2013-12-12 | 2013-12-12 | A kind of various dimensions credit management method based on distributed computing |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310681954.9A CN104717244B (en) | 2013-12-12 | 2013-12-12 | A kind of various dimensions credit management method based on distributed computing |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN104717244A true CN104717244A (en) | 2015-06-17 |
| CN104717244B CN104717244B (en) | 2018-12-14 |
Family
ID=53416204
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201310681954.9A Active CN104717244B (en) | 2013-12-12 | 2013-12-12 | A kind of various dimensions credit management method based on distributed computing |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104717244B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108364226A (en) * | 2018-01-26 | 2018-08-03 | 阿里巴巴集团控股有限公司 | The method and apparatus for identifying credible transaction |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20050165503A1 (en) * | 2004-01-28 | 2005-07-28 | Nec Corporation | Value information management system and method therefor |
| CN101685519A (en) * | 2008-09-22 | 2010-03-31 | 浙江大学 | Credit evaluation method and credit evaluation system |
| CN102222305A (en) * | 2011-06-13 | 2011-10-19 | 北京理工大学 | Creditworthiness evaluation method under distributed environment |
| CN102236851A (en) * | 2010-04-21 | 2011-11-09 | 百度在线网络技术（北京）有限公司 | Real-time computation method and system of multi-dimensional credit system based on user empowerment |
-
2013
- 2013-12-12 CN CN201310681954.9A patent/CN104717244B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20050165503A1 (en) * | 2004-01-28 | 2005-07-28 | Nec Corporation | Value information management system and method therefor |
| CN101685519A (en) * | 2008-09-22 | 2010-03-31 | 浙江大学 | Credit evaluation method and credit evaluation system |
| CN102236851A (en) * | 2010-04-21 | 2011-11-09 | 百度在线网络技术（北京）有限公司 | Real-time computation method and system of multi-dimensional credit system based on user empowerment |
| CN102222305A (en) * | 2011-06-13 | 2011-10-19 | 北京理工大学 | Creditworthiness evaluation method under distributed environment |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108364226A (en) * | 2018-01-26 | 2018-08-03 | 阿里巴巴集团控股有限公司 | The method and apparatus for identifying credible transaction |
| CN108364226B (en) * | 2018-01-26 | 2021-08-10 | 创新先进技术有限公司 | Method and device for identifying trusted transactions |
Also Published As
| Publication number | Publication date |
|---|---|
| CN104717244B (en) | 2018-12-14 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110135844B (en) | Credit recording and inquiring method and device based on block chain and electronic equipment | |
| US9635016B2 (en) | Cyber gene identification technology based on entity features in cyber space | |
| US10445734B2 (en) | Method and apparatus of identifying a transaction risk | |
| CN110827028B (en) | Block chain-based data acquisition transaction system and method | |
| CN102067126A (en) | Monetizing and prioritizing results of a distributed search | |
| AU2019101565A4 (en) | User data sharing method and device | |
| CN107294974A (en) | The method and apparatus for recognizing target clique | |
| Guo et al. | An Attribute-Based Trust Negotiation Protocol for D2D Communication in Smart City Balancing Trust and Privacy. | |
| CN106779608B (en) | Information processing method and information processing system based on WeChat platform | |
| CN105389488A (en) | Identity authentication method and apparatus | |
| CN104579909A (en) | Method and equipment for classifying user information and acquiring user grouping information | |
| CN110620820A (en) | Ubiquitous power Internet of things intelligent management system | |
| Lauw et al. | Stevent: Spatio-temporal event model for social network discovery | |
| CN111737765A (en) | Method and device for managing block chain resources | |
| CN111582935A (en) | Block chain-based integral mutual identification method and system | |
| CN108648017B (en) | User requirement matching method, device, equipment and storage medium easy to expand | |
| CN110297948B (en) | Relational network construction method and device | |
| CN104717244A (en) | Multidimensional credibility management method based on distributed computation | |
| CN114205816B (en) | Electric power mobile internet of things information security architecture and application method thereof | |
| CN111125747A (en) | Commodity browsing privacy protection method and system for commercial website user | |
| Ghiringhelli et al. | Recursive estimation of the spatial error model | |
| Zhang et al. | How do you interact with your old friends on a new site: Understanding social ties among different social network sites | |
| Razaque et al. | Fuzzy-gra trust model for cloud risk management | |
| Kpiebaareh et al. | User-connection behaviour analysis in service management using bipartite labelled property graph | |
| CN110414984A (en) | Auth method and Related product 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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |