CN107808324A - A kind of prestige value calculating method, network trading platform, the computer of online commodity - Google Patents
A kind of prestige value calculating method, network trading platform, the computer of online commodity Download PDFInfo
- Publication number
- CN107808324A CN107808324A CN201710996982.8A CN201710996982A CN107808324A CN 107808324 A CN107808324 A CN 107808324A CN 201710996982 A CN201710996982 A CN 201710996982A CN 107808324 A CN107808324 A CN 107808324A
- Authority
- CN
- China
- Prior art keywords
- mrow
- msub
- mover
- evaluation
- represent
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/06—Buying, selling or leasing transactions
- G06Q30/0601—Electronic shopping [e-shopping]
- G06Q30/0609—Buyer or seller confidence or verification
Landscapes
- Business, Economics & Management (AREA)
- Accounting & Taxation (AREA)
- Finance (AREA)
- Development Economics (AREA)
- Economics (AREA)
- Marketing (AREA)
- Strategic Management (AREA)
- Physics & Mathematics (AREA)
- General Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention belongs to technical field of electronic commerce, discloses a kind of prestige value calculating method, network trading platform, the computer of online commodity, including:Malice evaluation detection, user interest degree evaluation and test, credit value calculate.It is of the invention to utilize two-phase etection theory, the slight change of evaluation model is detected, situation when malicious user carries out malice evaluation to all commodity can be handled, while situation when malicious user only makes malice evaluation to end article can also be handled.The present invention determines the weight of different user evaluation by the similitude between comparison object item and user's scoring item, can more accurately reflect user's degree interested in end article and the reliability of user.
Description
Technical field
The invention belongs to technical field of electronic commerce, more particularly to a kind of prestige value calculating method, the network of online commodity
Transaction platform, computer.
Background technology
With the flourishing hair of ecommerce (network trading platform such as eBay, Amazon, Taobao, Youtube, Netflix)
Exhibition, according to the report of ecommerce foundation, exceed half in global most area is had online using the crowd of internet
The experience of shopping.Ecommerce has very big potentiality to promote economic development, is offered convenience to daily life.With network
Transaction platform provides increasing commodity and service, the range of choice of consumer are also increasing.But at the same time, occur with
Interests provide counterfeit goods for the seller being oriented to.For driving for economic interests, substantial amounts of seller is ready different to the product of oneself
Often high evaluation, and the evaluation to their rival's commodity is very low;Abnormal evaluation is referred to as malice and evaluated, and those are given
Malice pricer is given to be referred to as attacker.When attacker carries out malice to various commodity to be evaluated, service end is held very much
Their confidence level is easily determined according to their evaluation history and existing work, their malice can be effectively detected and comment
Valency.Once attacker only provides malice to end article (or a small group end article) and evaluated, and other commodity are only provided just
Often during evaluation, it is desirable to solution method in addition.When malice evaluation is intensity, it is meant that in a short time can be quickly
Ground provides a large amount of evaluations of end article credit value.Because evaluating data is all open to owner, attacker can monitor evaluation
Data, and provide malice and evaluate.So that the distribution that attacker is possible to manipulate malice evaluation follows any pattern, wherein the worst
Situation be malice evaluation be uniformly distributed between all evaluations.How the weight of different user evaluation is determined;Weight reflects
The reliability of user.But different users can have different preferences, the interest of a user is generally focused on some themes
Or in classification, without being dispersed on all commodity;The weight of evaluation should reflect that user is interested in end article
Degree.Patent application " a kind of dynamic Web service trust evaluation method based on user's honesty " (Shen that Southeast China University proposes
Please number B of 2012102796.4 application publication number CN 102880637) disclose a kind of dynamic Web based on user's honesty
Service trust appraisal procedure, is comprised the concrete steps that:First according to requirement of the user to service, automatic mining user preference;According to
Family preference division similar users group, carries out evaluating uniformity cluster in preference similar users group, distinguishes honest user and malice
User, user's honesty of preference similar users group is calculated, according to the change of user's honesty, dynamic adjusts subjective assessment comprehensive
The weight trusted in calculating is closed, so as to reduce influence of the malice evaluation to Web service trust evaluation, improves the accurate of trust evaluation
Property, preferably instruct user to carry out the selection of trusted service.Deficiency existing for the patent application is:Do not consider how that processing is disliked
User anticipate only to end article, rather than all commodity are made with situation when malice is evaluated, once malicious user is only to target
Commodity (or a small group end article) provide malice and evaluated, and when only providing normal evaluation to other commodity, this model just can not
The slight change of evaluation model is perceived, credit value assessment intactly can not be carried out to commodity, mistake is carried out so as to mislead user
Decision-making by mistake.Patent application " credit computing method in a kind of credit system " (Shen that Acoustical Inst., Chinese Academy of Sciences proposes
Please number B of 200710122393.3 application publication number CN 10199683) a kind of credit computing method in credit system is disclosed,
Comprise the concrete steps that:Node i sends request to the surroundings nodes for the node j that need to calculate prestige, and inquiry surroundings nodes are commented node j
Valency;Surroundings nodes return to evaluation;According to the reliability of the irrelevance of each node and irrelevance, node i amendment evaluation node is given
The evaluation gone out;Node i merges oneself with the evaluation of revised evaluation node with service node j practical experience, calculates
Last prestige.Deficiency existing for the patent application is:Do not consider how to determine different nodes (i.e. not in this method model
Same user) evaluation weight, the weight of evaluation can reflect that user's degree interested in end article and user's is reliable
Property.Evaluation of the user to service has much relations with its preference, if the weight of different user evaluation can not be determined, then the
The evaluation of tripartite or objective capability confidence level can not cover the user of difference preference, so as to produce misjudgment phenomenon.
In summary, the problem of prior art is present be:Current credit value technical method, which exists, not to be considered how to locate
Malicious user is managed only to end article, rather than all commodity are made with situation when malice is evaluated,;Do not consider how to determine
The weight of different Node evaluations, the weight of evaluation can not can reflect user's degree interested in end article and user
Reliability.
The content of the invention
The problem of existing for prior art, the invention provides a kind of prestige value calculating method, the network of online commodity
Transaction platform, computer.
The present invention is achieved in that a kind of prestige value calculating method of online commodity, including server carry out malice and commented
Valency detects and user interest degree evaluation and test, and finally comprehensive malice evaluation testing result and user interest degree carry out the credit value meter of commodity
Calculate.The prestige value calculating method of the online commodity comprises the following steps:
Step 1, malice evaluation detection:
(1) server is receiving user's requirement calculating end article itDuring the request of credit value, it will be collected on target
Commodity itEvaluation of estimate arranged in chronological order, be one group per n evaluation of estimate;
(2) evaluation of estimate of kth group is expressed as
Wherein,J-th of evaluation of estimate in kth group is represented, n represents the group number of evaluation of estimate;
(3) evaluation of estimate will be providedUser's set representations be
Wherein,Expression provides evaluation of estimateUser collection,Expression provides evaluation of estimateUser;
(4) intermediate vector is constructed
Wherein,Represent intermediate vector,Represent j-th of evaluation of estimate in kth group, μk-1Represent that deleting malice evaluates it
The average value evaluated afterwards in (k-1) group, as k=1, μ0Represent the average value of evaluation of estimate in first packet, N+For positive integer
Symbol;
(5) detected using following formulaIn with the presence or absence of malice evaluate:
Wherein, n represents the group number of evaluation of estimate,Represent intermediate vectorAverage, s2Represent sample variance,Represent false
If the method for inspection, if inequality is set up,It is middle malice evaluation to be present, perform (6), conversely, thenIn be not present malice evaluate,
Perform step (7);
(6) calculated using following formulaAverage
Wherein,RepresentAverage,RepresentIn evaluation average, μk-1Represent to delete (k- after malice is evaluated
1) average value evaluated in group;
(7) log-likelihood ratio C is calculated using following formulaj:
Wherein, CjRepresent the log-likelihood ratio of j-th of evaluation of estimate, C0=0,RepresentAverage, σ2RepresentMiddle normal state
The variance of distribution,J-th of intermediate vector in kth group is represented,Represent intermediate vectorAverage;Work as CjWhen >=0.8,
It is judged as exceptional value,Malice is judged as to evaluate,Be judged as malicious user, and by its fromMiddle deletion, is obtained pair
Answer the normal users set of k-th of evaluation of estimate packetPerform step 2;
(8) if (5) are detectedIn be not present malice evaluate, then
Step 2, user interest degree evaluation and test:
(1) similarity is calculated using following formula
Wherein,Represent the similarity between end article and commodity in addition to end article, itRepresent target business
Product, i represent the commodity in addition to end article,A group of labels are represented, θ is representedIn some label, ∑ for plus and symbol,
Intensity is applied in rel expressions, and scope 0 to 1,0 represents do not have intensity completely, and 1 represents maximum intensity;
(2) user u interest-degree d is calculated using following formulau:
Wherein, duRepresent the interest-degree of user, ncRepresent end article itNumber,Represent what user u had been evaluated
Commodity number, ∑ represents to add and symbol,The similarity between end article and commodity in addition to end article is represented,Represent setIn element number;
Step 3, credit value calculate:
(1) end article i is calculated using following formulatCredit value:
Wherein, rep represents the credit value of commodity, itEnd article is represented, k represents kth group evaluation of estimate, and δ represents to comment recently
Intensity is applied in the weight evaluated in valency group, rel expressions,Expression deletes the user u of malicious user set, duRepresent user
Interest-degree,Represent user u for end article itSatisfaction, scope between 1 to 5,1 represent it is very dissatisfied, 5
Represent very satisfied;
(2) by end article itCredit value return to asked user u.
Further, the hypothesis testing methodRefer to examine the population variance of two normal random variables whether equal
A kind of hypothesis testing method,It is 1 to represent first free degree, and second free degree is n-1, and confidence level is α'sTest value,
F- distribution tables can be looked into obtain.
Further, the s2WithCircular it is as follows:
Wherein, s2Sample variance is represented, n represents the group number of evaluation of estimate,J-th of intermediate vector in kth group is represented,
Represent intermediate vectorAverage.
Further, the CjLog-likelihood ratio refer to:
Wherein,Withθ is represented respectively0And θ1Under probability density function, yjRepresent to be generated by the random process of parameter θ
A series of samples, N+For positive integer symbol.
Further, the σ2Circular it is as follows:
Wherein, σ2RepresentThe variance of middle normal distribution, n represent the group number of evaluation of estimate, and j represents the sequence number of evaluation of estimate, ∑
To add and symbol,J-th of intermediate vector in kth group is represented,Represent intermediate vectorAverage.
Another object of the present invention is to provide a kind of network friendship of the prestige value calculating method using the online commodity
Easy platform.
Another object of the present invention is to provide a kind of computer of the prestige value calculating method using the online commodity.
Another object of the present invention is to provide a kind of server of the prestige value calculating method using the online commodity.
Another object of the present invention is to provide a kind of computer of the prestige value calculating method using the online commodity
Program.
The present invention proposes a prestige Computational frame, to obtain the reliable credit value of any online commodity to help consumer
Make correct purchase decision.Three modules are included in the frame:Malice grading detection, interest-degree calculate and prestige calculates mould
Block.Collected by data collector, malice is evaluated detector and can evaluated by two-phase detection method to detect malice.Experimental data
Show, for prior art when not filtering out malice and evaluating, mean absolute error between result of calculation and ideal value can be with
The increase of ratio of the malice evaluation in all evaluations and it is linearly increasing.The present invention can be by the mean absolute error of result of calculation
Maintain near 0, show the very close ideal value of credit value of calculating and can effectively handle equally distributed malice evaluation.
The present invention determines the weight of different user evaluation by the similitude between comparison object item and user's scoring item, can be more accurate
Really reflect user's degree interested in end article and the reliability of user.
Brief description of the drawings
Fig. 1 is the prestige value calculating method flow chart of online commodity provided in an embodiment of the present invention.
Embodiment
In order to make the purpose , technical scheme and advantage of the present invention be clearer, with reference to embodiments, to the present invention
It is further elaborated.It should be appreciated that the specific embodiments described herein are merely illustrative of the present invention, it is not used to
Limit the present invention.
Ecommerce has the experience of online shopping in global most area more than half using the crowd of internet;Electricity
Sub- commercial affairs have very big potentiality to promote economic development, are offered convenience to daily life.
The application principle of the present invention is explained in detail below in conjunction with the accompanying drawings.
As shown in figure 1, the prestige value calculating method of online commodity provided in an embodiment of the present invention comprises the following steps:
S101:Malice evaluation detection;
S102:User interest degree is evaluated and tested;
S103:Credit value calculates.
The application principle of the present invention is further described with reference to specific embodiment.
The prestige value calculating method of online commodity provided in an embodiment of the present invention comprises the following steps:
Step 1, malice evaluation detection;
(1) server is receiving user's requirement calculating end article itDuring the request of credit value, it will be collected on target
Commodity itEvaluation of estimate arranged in chronological order, be one group per n evaluation of estimate.Server represents the evaluation of estimate of kth group
ForTo embody method as follows:
Wherein,J-th of evaluation of estimate in kth group is represented, n represents the group number of evaluation of estimate;
(2) server will provide evaluation of estimateUser's set representations be
Wherein,Expression provides evaluation of estimateUser collection,Expression provides evaluation of estimateUser;
(3) intermediate vector is constructedWillIt is converted into
Wherein,Represent intermediate vector,Represent j-th of evaluation of estimate in kth group, μk-1Represent that deleting malice evaluates it
The average value evaluated afterwards in (k-1) group, as k=1, μ0Represent the average value of evaluation of estimate in first packet, N+For positive integer
Symbol;
(4) server by utilizing following formula detectsIn with the presence or absence of malice evaluate:
Wherein, n represents the group number of evaluation of estimate,Represent intermediate vectorAverage,s2Represent sample side
Difference,Represent hypothesis testing method, it is assumed that the method for inspectionRefer to examine two normal states random
A kind of whether equal hypothesis testing method of the population variance of variable,It is 1 to represent first free degree, and second free degree is
N-1, confidence level are α'sTest value, F- distribution tables can be looked into and obtained.If inequality is set up,It is middle malice evaluation to be present, perform
(6), conversely, thenIn be not present malice evaluate, perform (7);
(5) server by utilizing following formula calculatesAverage
Wherein,RepresentAverage,RepresentIn evaluation average, μk-1Represent to delete (k- after malice is evaluated
1) average value evaluated in group;
(6) server by utilizing following formula calculates log-likelihood ratio Cj:
Wherein, C0=0,RepresentAverage, σ2RepresentThe variance of middle normal distribution, Table
Show j-th of intermediate vector in kth group,Represent intermediate vectorAverage;Work as CjWhen >=0.8,It is judged as exceptional value,Malice is judged as to evaluate,Be judged as malicious user, and by its fromMiddle deletion, obtain corresponding k-th of evaluation of estimate point
The normal users set of groupPerform step 2;
C in above formulajRepresent the log-likelihood ratio of j-th of evaluation of estimate, CjLog-likelihood ratio circular it is as follows:
Wherein,Withθ is represented respectively0And θ1Under probability density function, yjRepresent to be generated by the random process of parameter θ
A series of samples, N+For positive integer symbol;
(7) if (4) are detectedIn be not present malice evaluate, then
Step 2:User interest degree is evaluated and tested.
(1) server by utilizing following formula calculates similarity
Wherein,Represent the similarity between end article and commodity in addition to end article, itRepresent target business
Product, i represent the commodity in addition to end article,A group of labels are represented, θ is representedIn some label, ∑ for plus and symbol,
Intensity is applied in rel expressions, and scope is between 0 to 1, and 0 represents do not have intensity completely, and 1 represents maximum intensity;
(2) server by utilizing following formula calculates user u interest-degree du:
Wherein, duRepresent the interest-degree of user, ncRepresent end article itNumber,Represent what user u had been evaluated
Commodity number, ∑ represents to add and symbol,The similarity between end article and commodity in addition to end article is represented,Represent setIn element number.
Step 3:Credit value calculates:
(1) server by utilizing following formula calculates end article itCredit value:
Wherein, rep represents the credit value of commodity, itEnd article is represented, k represents kth group evaluation of estimate, and δ represents to comment recently
Intensity is applied in the weight evaluated in valency group, rel expressions,Expression deletes the user u of malicious user set, duRepresent user
Interest-degree,Represent user u for end article itSatisfaction, scope between 1 to 5,1 represent it is very dissatisfied, 5
Represent very satisfied;
(2) server is by end article itCredit value return to asked user u.
The foregoing is merely illustrative of the preferred embodiments of the present invention, is not intended to limit the invention, all essences in the present invention
All any modification, equivalent and improvement made within refreshing and principle etc., should be included in the scope of the protection.
Claims (9)
- A kind of 1. prestige value calculating method of online commodity, it is characterised in that the prestige value calculating method bag of the online commodity Include following steps:Step 1, malice evaluation detection:(1) server is receiving user's requirement calculating end article itDuring the request of credit value, it will be collected on end article it Evaluation of estimate arranged in chronological order, be one group per n evaluation of estimate;(2) evaluation of estimate of kth group is expressed as<mrow> <mover> <msub> <mi>y</mi> <mi>k</mi> </msub> <mo>&RightArrow;</mo> </mover> <mo>=</mo> <mo>{</mo> <mover> <msubsup> <mi>y</mi> <mi>k</mi> <mi>j</mi> </msubsup> <mo>&RightArrow;</mo> </mover> <mo>:</mo> <mi>j</mi> <mo>=</mo> <mn>1</mn> <mo>,</mo> <mn>2</mn> <mo>,</mo> <mo>...</mo> <mo>,</mo> <mi>n</mi> <mo>}</mo> <mo>,</mo> <mi>n</mi> <mo>,</mo> <mi>k</mi> <mo>&Element;</mo> <msup> <mi>N</mi> <mo>+</mo> </msup> <mo>;</mo> </mrow>Wherein,J-th of evaluation of estimate in kth group is represented, n represents the group number of evaluation of estimate;(3) evaluation of estimate will be providedUser's set representations be<mrow> <mover> <msub> <mi>x</mi> <mi>k</mi> </msub> <mo>&RightArrow;</mo> </mover> <mo>=</mo> <mo>{</mo> <mover> <msubsup> <mi>x</mi> <mi>k</mi> <mi>j</mi> </msubsup> <mo>&RightArrow;</mo> </mover> <mo>:</mo> <mi>j</mi> <mo>=</mo> <mn>1</mn> <mo>,</mo> <mn>2</mn> <mo>,</mo> <mo>...</mo> <mo>,</mo> <mi>n</mi> <mo>}</mo> <mo>,</mo> <mi>n</mi> <mo>&Element;</mo> <msup> <mi>N</mi> <mo>+</mo> </msup> <mo>;</mo> </mrow>Wherein,Expression provides evaluation of estimateUser collection,Expression provides evaluation of estimateUser;(4) intermediate vector is constructed<mrow> <mover> <msub> <mi>z</mi> <mi>k</mi> </msub> <mo>&RightArrow;</mo> </mover> <mo>=</mo> <mo>{</mo> <mover> <msubsup> <mi>y</mi> <mi>k</mi> <mi>j</mi> </msubsup> <mo>&RightArrow;</mo> </mover> <mo>-</mo> <msub> <mi>&mu;</mi> <mrow> <mi>k</mi> <mo>-</mo> <mn>1</mn> </mrow> </msub> <mo>:</mo> <mi>j</mi> <mo>=</mo> <mn>1</mn> <mo>,</mo> <mn>2</mn> <mo>,</mo> <mo>...</mo> <mo>,</mo> <mi>n</mi> <mo>}</mo> <mo>,</mo> <mi>n</mi> <mo>&Element;</mo> <msup> <mi>N</mi> <mo>+</mo> </msup> <mo>;</mo> </mrow>Wherein,Represent intermediate vector,Represent j-th of evaluation of estimate in kth group, μk-1Represent to delete the after malice is evaluated (k-1) average value evaluated in group, as k=1, μ0Represent the average value of evaluation of estimate in first packet, N+Accorded with for positive integer Number;(5) detected using following formulaIn with the presence or absence of malice evaluate:Wherein, n represents the group number of evaluation of estimate,Represent intermediate vectorAverage, s2Represent sample variance,Represent to assume inspection Proved recipe method, if inequality is set up,It is middle malice evaluation to be present, perform (6), conversely, thenIn be not present malice evaluate, hold Row step (8);(6) calculated using following formulaAverage<mrow> <mover> <msub> <mi>z</mi> <mi>k</mi> </msub> <mo>&OverBar;</mo> </mover> <mo>=</mo> <msub> <mover> <mi>y</mi> <mo>&OverBar;</mo> </mover> <mi>k</mi> </msub> <mo>-</mo> <msub> <mi>&mu;</mi> <mrow> <mi>k</mi> <mo>-</mo> <mn>1</mn> </mrow> </msub> <mo>;</mo> </mrow>Wherein,RepresentAverage,RepresentIn evaluation average, μk-1Represent to delete (k-1) group after malice is evaluated The average value of middle evaluation;(7) log-likelihood ratio C is calculated using following formulaj:<mrow> <msub> <mi>C</mi> <mi>j</mi> </msub> <mo>=</mo> <mfenced open = "{" close = ""> <mtable> <mtr> <mtd> <mrow> <msub> <mi>C</mi> <mrow> <mi>j</mi> <mo>-</mo> <mn>1</mn> </mrow> </msub> <mo>+</mo> <mfrac> <mover> <msub> <mi>z</mi> <mi>k</mi> </msub> <mo>&OverBar;</mo> </mover> <msup> <mi>&sigma;</mi> <mn>2</mn> </msup> </mfrac> <mrow> <mo>(</mo> <mover> <msubsup> <mi>z</mi> <mi>k</mi> <mi>j</mi> </msubsup> <mo>&RightArrow;</mo> </mover> <mo>-</mo> <mfrac> <mover> <msub> <mi>z</mi> <mi>k</mi> </msub> <mo>&OverBar;</mo> </mover> <mn>2</mn> </mfrac> <mo>)</mo> </mrow> <mo>,</mo> <msub> <mi>C</mi> <mrow> <mi>j</mi> <mo>-</mo> <mn>1</mn> </mrow> </msub> <mo><</mo> <mn>0.8</mn> </mrow> </mtd> </mtr> <mtr> <mtd> <mrow> <mfrac> <mover> <msub> <mi>z</mi> <mi>k</mi> </msub> <mo>&OverBar;</mo> </mover> <msup> <mi>&sigma;</mi> <mn>2</mn> </msup> </mfrac> <mrow> <mo>(</mo> <mover> <msubsup> <mi>z</mi> <mi>k</mi> <mi>j</mi> </msubsup> <mo>&RightArrow;</mo> </mover> <mo>-</mo> <mfrac> <mover> <msub> <mi>z</mi> <mi>k</mi> </msub> <mo>&OverBar;</mo> </mover> <mn>2</mn> </mfrac> <mo>)</mo> </mrow> <mo>,</mo> <msub> <mi>C</mi> <mrow> <mi>j</mi> <mo>-</mo> <mn>1</mn> </mrow> </msub> <mo>&GreaterEqual;</mo> <mn>0.8</mn> </mrow> </mtd> </mtr> </mtable> </mfenced> <mo>;</mo> </mrow>Wherein, CjRepresent the log-likelihood ratio of j-th of evaluation of estimate, C0=0,RepresentAverage, σ2RepresentMiddle normal distribution Variance,J-th of intermediate vector in kth group is represented,Represent intermediate vectorAverage;Work as CjWhen >=0.8,It is judged to It is set to exceptional value,Malice is judged as to evaluate,Be judged as malicious user, and by its fromMiddle deletion, obtain corresponding kth The normal users set of individual evaluation of estimate packetPerform step 2;(8) if (5) are detectedIn be not present malice evaluate, thenStep 2, user interest degree evaluation and test:(1) similarity is calculated using following formulaWherein,Represent the similarity between end article and commodity in addition to end article, itRepresent end article, i The commodity in addition to end article are represented,A group of labels are represented, θ is representedIn some label, ∑ for plus and symbol, rel Intensity is applied in expression, and scope 0 to 1,0 represents do not have intensity completely, and 1 represents maximum intensity;(2) user u interest-degree d is calculated using following formulau:Wherein, duRepresent the interest-degree of user, ncRepresent end article itNumber,Represent the commodity that user u had been evaluated Number, ∑ represents to add and symbol,The similarity between end article and commodity in addition to end article is represented,Table Show setIn element number;Step 3, credit value calculate:(1) end article i is calculated using following formulatCredit value:Wherein, rep represents the credit value of commodity, itEnd article is represented, k represents kth group evaluation of estimate, and δ is represented in nearest evaluation group Intensity is applied in the weight of evaluation, rel expressions,Expression deletes the user u of malicious user set, duRepresent the interest of user Degree,Represent user u for end article itSatisfaction, scope between 1 to 5,1 represent it is very dissatisfied, 5 represent it is non- It is often satisfied;(2) by end article itCredit value return to asked user u.
- 2. the prestige value calculating method of online commodity as claimed in claim 1, it is characterised in that the hypothesis testing method Refer to a kind of hypothesis testing method for examining the population variance of two normal random variables whether equal,Represent first certainly By spending for 1, second free degree is n-1, and confidence level is α'sTest value, F- distribution tables can be looked into and obtained.
- 3. the prestige value calculating method of online commodity as claimed in claim 1, it is characterised in that the s2WithSpecific meter Calculation method is as follows:<mrow> <msup> <mi>s</mi> <mn>2</mn> </msup> <mo>=</mo> <mfrac> <mn>1</mn> <mrow> <mi>n</mi> <mo>-</mo> <mn>1</mn> </mrow> </mfrac> <munderover> <mo>&Sigma;</mo> <mrow> <mi>j</mi> <mo>=</mo> <mn>1</mn> </mrow> <mi>n</mi> </munderover> <msup> <mrow> <mo>(</mo> <mover> <msubsup> <mi>z</mi> <mi>k</mi> <mi>j</mi> </msubsup> <mo>&RightArrow;</mo> </mover> <mo>-</mo> <msub> <mover> <mi>z</mi> <mo>&OverBar;</mo> </mover> <mi>k</mi> </msub> <mo>)</mo> </mrow> <mn>2</mn> </msup> <mo>,</mo> <msub> <mover> <mi>z</mi> <mo>&OverBar;</mo> </mover> <mi>k</mi> </msub> <mo>=</mo> <mfrac> <mn>1</mn> <mi>n</mi> </mfrac> <munderover> <mo>&Sigma;</mo> <mrow> <mi>j</mi> <mo>=</mo> <mn>1</mn> </mrow> <mi>n</mi> </munderover> <mover> <msubsup> <mi>z</mi> <mi>k</mi> <mi>j</mi> </msubsup> <mo>&RightArrow;</mo> </mover> <mo>;</mo> </mrow>Wherein, s2Sample variance is represented, n represents the group number of evaluation of estimate,J-th of intermediate vector in kth group is represented,Represent Intermediate vectorAverage.
- 4. the prestige value calculating method of online commodity as claimed in claim 1, it is characterised in that the CjLog-likelihood ratio Refer to:<mrow> <msub> <mi>C</mi> <mi>j</mi> </msub> <mo>=</mo> <mi>l</mi> <mi>n</mi> <mfrac> <mrow> <msub> <mi>p</mi> <msub> <mi>&theta;</mi> <mn>1</mn> </msub> </msub> <mrow> <mo>(</mo> <msub> <mi>y</mi> <mi>j</mi> </msub> <mo>)</mo> </mrow> </mrow> <mrow> <msub> <mi>p</mi> <msub> <mi>&theta;</mi> <mn>0</mn> </msub> </msub> <mrow> <mo>(</mo> <msub> <mi>y</mi> <mi>j</mi> </msub> <mo>)</mo> </mrow> </mrow> </mfrac> <mo>,</mo> <mi>j</mi> <mo>&Element;</mo> <msup> <mi>N</mi> <mo>+</mo> </msup> <mo>;</mo> </mrow>Wherein,Withθ is represented respectively0And θ1Under probability density function, yjRepresent one generated by the random process of parameter θ Serial sample, N+For positive integer symbol.
- 5. the prestige value calculating method of online commodity as claimed in claim 1, it is characterised in that the σ2Specific calculating side Method is as follows:<mrow> <msup> <mi>&sigma;</mi> <mn>2</mn> </msup> <mo>=</mo> <mfrac> <mn>1</mn> <mi>n</mi> </mfrac> <munderover> <mo>&Sigma;</mo> <mrow> <mi>j</mi> <mo>=</mo> <mn>1</mn> </mrow> <mi>n</mi> </munderover> <msup> <mrow> <mo>(</mo> <mover> <msubsup> <mi>z</mi> <mi>k</mi> <mi>j</mi> </msubsup> <mo>&RightArrow;</mo> </mover> <mo>-</mo> <msub> <mover> <mi>z</mi> <mo>&OverBar;</mo> </mover> <mi>k</mi> </msub> <mo>)</mo> </mrow> <mn>2</mn> </msup> <mo>;</mo> </mrow>Wherein, σ2RepresentThe variance of middle normal distribution, n represent evaluation of estimate group number, j represent evaluation of estimate sequence number, ∑ for plus and Symbol,J-th of intermediate vector in kth group is represented,Represent intermediate vectorAverage.
- A kind of 6. network trading platform using the prestige value calculating method of online commodity described in any one of Claims 1 to 55.
- A kind of 7. computer using the prestige value calculating method of online commodity described in any one of Claims 1 to 55.
- A kind of 8. server using the prestige value calculating method of online commodity described in any one of Claims 1 to 55.
- A kind of 9. computer program using the prestige value calculating method of online commodity described in any one of Claims 1 to 55.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710996982.8A CN107808324B (en) | 2017-10-20 | 2017-10-20 | Online commodity credit value calculation method, network transaction platform and computer |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710996982.8A CN107808324B (en) | 2017-10-20 | 2017-10-20 | Online commodity credit value calculation method, network transaction platform and computer |
Publications (2)
Publication Number | Publication Date |
---|---|
CN107808324A true CN107808324A (en) | 2018-03-16 |
CN107808324B CN107808324B (en) | 2021-05-28 |
Family
ID=61592358
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201710996982.8A Active CN107808324B (en) | 2017-10-20 | 2017-10-20 | Online commodity credit value calculation method, network transaction platform and computer |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN107808324B (en) |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108665339A (en) * | 2018-03-27 | 2018-10-16 | 北京航空航天大学 | A kind of electric business product reliability index and its implementation estimated based on subjective emotion |
CN111915210A (en) * | 2020-08-11 | 2020-11-10 | 曲阜师范大学 | Trust management method in distributed Internet of things based on block chain |
Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103957062A (en) * | 2014-03-31 | 2014-07-30 | 西安电子科技大学 | Method for evaluating reputation of cognitive users in distributed cognitive radio network |
CN104350490A (en) * | 2012-04-11 | 2015-02-11 | 新加坡国立大学 | Methods, apparatuses and computer-readable mediums for organizing data relating to a product |
US20160034968A1 (en) * | 2014-07-31 | 2016-02-04 | Huawei Technologies Co., Ltd. | Method and device for determining target user, and network server |
CN106296282A (en) * | 2016-08-08 | 2017-01-04 | 南京大学 | A kind of net purchase Product evaluation method marked based on user comment and history |
CN106570525A (en) * | 2016-10-26 | 2017-04-19 | 昆明理工大学 | Method for evaluating online commodity assessment quality based on Bayesian network |
CN106779858A (en) * | 2016-12-26 | 2017-05-31 | 西安理工大学 | A kind of product analysis method based on multidimensional perception information semantic level association |
-
2017
- 2017-10-20 CN CN201710996982.8A patent/CN107808324B/en active Active
Patent Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104350490A (en) * | 2012-04-11 | 2015-02-11 | 新加坡国立大学 | Methods, apparatuses and computer-readable mediums for organizing data relating to a product |
CN103957062A (en) * | 2014-03-31 | 2014-07-30 | 西安电子科技大学 | Method for evaluating reputation of cognitive users in distributed cognitive radio network |
US20160034968A1 (en) * | 2014-07-31 | 2016-02-04 | Huawei Technologies Co., Ltd. | Method and device for determining target user, and network server |
CN106296282A (en) * | 2016-08-08 | 2017-01-04 | 南京大学 | A kind of net purchase Product evaluation method marked based on user comment and history |
CN106570525A (en) * | 2016-10-26 | 2017-04-19 | 昆明理工大学 | Method for evaluating online commodity assessment quality based on Bayesian network |
CN106779858A (en) * | 2016-12-26 | 2017-05-31 | 西安理工大学 | A kind of product analysis method based on multidimensional perception information semantic level association |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108665339A (en) * | 2018-03-27 | 2018-10-16 | 北京航空航天大学 | A kind of electric business product reliability index and its implementation estimated based on subjective emotion |
CN111915210A (en) * | 2020-08-11 | 2020-11-10 | 曲阜师范大学 | Trust management method in distributed Internet of things based on block chain |
Also Published As
Publication number | Publication date |
---|---|
CN107808324B (en) | 2021-05-28 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Zhang et al. | Gcn-based user representation learning for unifying robust recommendation and fraudster detection | |
Abdel-Basset et al. | A group decision making framework based on neutrosophic VIKOR approach for e-government website evaluation | |
Liu et al. | Holoscope: Topology-and-spike aware fraud detection | |
Li et al. | Robust reputation-based ranking on bipartite rating networks | |
Costa et al. | Detecting tip spam in location-based social networks | |
Lim et al. | A topological approach for detecting twitter communities with common interests | |
Jang et al. | Distance-based customer detection in fake follower markets | |
Yu et al. | Hybrid attacks on model-based social recommender systems | |
Pham et al. | Sentiment analysis and user similarity for social recommender system: An experimental study | |
Esslimani et al. | Detecting leaders in behavioral networks | |
Gong | A Personalized Recommendation Algorithm on Integration of Item Semantic Similarity and Item Rating Similarity. | |
CN107808324A (en) | A kind of prestige value calculating method, network trading platform, the computer of online commodity | |
Ebrahimi et al. | Personalized recommender system based on social relations | |
de Bruin et al. | Experimental evaluation of train and test split strategies in link prediction | |
Rajeh et al. | Comparing community-aware centrality measures in online social networks | |
Setiawan et al. | Preliminary study of spam profile detection for social media using Markov clustering: Case study on Javanese people | |
Krol et al. | On rumor source detection and its experimental verification on twitter | |
Miao et al. | Identifying the most influential topic-sensitive opinion leaders in online review communities | |
Nabipourshiri et al. | Tree-based classification to users' trustworthiness in OSNs | |
Nakatsuji et al. | Recommendations over domain specific user graphs | |
US20220360604A1 (en) | System and method for social network analysis | |
Hsu et al. | Using the beta distribution technique to detect attacked items from collaborative filtering | |
Agarwal et al. | Using trust model for detecting malicious activities in Twitter | |
Mohammadinejad et al. | Employing personality feature to rank the influential users in signed networks | |
Xiao et al. | A novel trust evaluation mechanism for collaborative filtering recommender systems |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |