CN104766215A

CN104766215A - Comprehensive multi-dimension goods owner selection quantification method

Info

Publication number: CN104766215A
Application number: CN201510173407.9A
Authority: CN
Inventors: 李敬泉; 吴广盛; 方慧敏; 陈威
Original assignee: Nanjing University
Current assignee: Zhongchu Zhiyun Technology Co ltd
Priority date: 2015-04-13
Filing date: 2015-04-13
Publication date: 2015-07-08
Anticipated expiration: 2035-04-13
Also published as: CN104766215B

Abstract

The invention discloses a comprehensive multi-dimension goods owner selection quantification method enabling a member to evaluate a goods owner objectively and comprehensively from multiple dimensions in a quantitative mode. According to the method, multiple measure indexes affecting selection including loading and unloading punctuality, the number of times of carriage, the number of times of complaints from platform members, goods payment and information issuing timeliness ratio, and one-time offer close deal rate are fully considered, the weight of each index is determined quantitatively based on the index weight division method of a multi-attribute decision-making model combining the least square method and the ELECTRE-II method, reasonable evaluation of the credibility of a goods owner on an e-commerce platform with logistics service as transaction object is well achieved through precise big data mining and analyzing and model algorithms, and platform members can judge the comprehensive condition of a goods owner and make choices meeting their own needs.

Description

Comprehensive and multidimensional owner selection quantification method

Technical Field

The invention relates to a credibility quantifying and comprehensive evaluating system for a shipper side in a logistics electronic commerce platform, which is suitable for objective and comprehensive evaluation and quantification of the shipper side from multiple dimensions and belongs to the technical field of data analysis.

Background

The continuous development of information technology has promoted the deep integration of electronic commerce and logistics industry, however, the non-face-to-face transaction brings some potential risks while creating a new business model. In the process of distinguishing, detecting and managing risks, a credit security mechanism is established, so that a user can integrate credibility of owners, platform members and the like and select proper owners, and therefore the method is very important for saving transportation cost and improving logistics efficiency.

Most of the existing logistics e-commerce platforms are based on the credit evaluation method of the commodity e-commerce platform, the establishment of an evaluation index system is relatively unsound, the real bearing level of a goods owner cannot be objectively and fully reflected, and other users of the logistics e-commerce platform cannot know the real and reliable goods owner bearing condition.

Disclosure of Invention

The purpose of the invention is as follows: aiming at the problems and the defects in the existing logistics e-commerce platform evaluation system, the invention provides a goods owner bearing level credibility degree quantification and comprehensive evaluation selection method based on a combined least square method and a multi-attribute decision model on the basis of all historical evaluation and transaction record data in a transaction platform.

The technical scheme is as follows: a comprehensive and multidimensional owner selection quantification method is suitable for quantifying and selecting owners in an electronic commerce platform providing logistics transaction services. The method specifically comprises the following steps:

(1) selecting credibility evaluation scheme attribute, constructing scheme set and attribute set by data analysis

And selecting credibility evaluation scheme attributes required by the platform by referring to the database and the industry indexes, and establishing a scheme set and an attribute set according to the credibility evaluation scheme and the mutual relation among the attributes. Credibility evaluation scheme set: the credibility of the same industry, the credibility of the goods to be evaluated and the credibility of the platform reference. Attribute set: seven evaluation indexes, namely punctual shipment of a shipper, punctual discharge of a consignee, the number of times of a shipper service, the number of times of complaints of members of a subject platform, the information release timeliness rate, the payment timeliness rate of the shipper payment of the shipper and the one-time quoted price and order rate, are selected as attributes influencing the scheme.

(2) Attribute set normalization process

According to the influence of the attribute on the credibility evaluation scheme, the attribute of the scheme is beneficial and cost-effective. The larger the attribute value of the benefit attribute is, the better, whereas the smaller the attribute value of the cost attribute is, the better. And the dimension and dimension unit of each scheme attribute are different, so each attribute of the scheme needs to be subjected to non-dimensionalization processing, and the non-dimensionalization processing of the benefit attribute and the cost attribute is as follows:

and (4) benefit type attribute processing: b_ij＝(a_ij-a_j ^min)/(a_j ^max-a_j ^min)

Cost type attribute processing: b_ij＝(a_j ^max-a_ij)/(a_j ^max-a_j ^min)

Wherein, a_ijIs the attribute value of the jth attribute of the scheme i, b_ijIs a_ijNormalized value of a_j ^maxIs the jth attribute P_jMaximum value of a_j ^minIs P_jIs measured. b_ijE (0,1), i ═ 1, 2., m, j ═ 1, 2., n, normalized matrix B ═ B_ij)_m×n。

(3) Performing multidimensional analysis and constructing attribute weight multivariate optimization model

Considering the effectiveness of information feedback and the reasonability of information processing, a multivariate optimization model is introduced, different influences of the multivariate optimization model on the credibility of the shipper are comprehensively considered from two different dimensions of the shipper members and the shipper, and the multivariate optimization model is numerically depicted on the shipper members and the shipper on the basis of big data analysis. The contact points between the carrier members and the platform and the shipper are different, and therefore different attribute evaluation values are shown for the emphasis on the reliability of the shipper.

(ii) multivariate weight determination

In order to reduce the influence of single preference and cognitive limitation on results, the platform carries out weighted average on historical evaluation scores of all the carrier members on each index of the shipper to obtain the most intuitive comprehensive evaluation score of the carrier members on the shipper; in order to reduce the deviation caused by evaluation of a single dimension, a platform decision element is introduced from another dimension, and the credibility of a buyer is subjected to supplementary evaluation scoring based on the historical trading data and trading behaviors of the platform. And then, carrying out standardization processing on the scores to obtain the weight value of each attribute, and finally setting an importance coefficient for the two decision elements based on the integrity and the authenticity of the data.

Decision element T determined by two dimensions_kK is 1, 2; wherein T is₁Representative Carrier Member decision element, T₂Representing platform decision elements.

The attribute weights assigned by the two decision elements are: w is a^k＝(w₁ ^k,w₂ ^k,...,w_n ^k)^T，k＝1,2

The importance degree of each decision element is as follows: z is (z)₁,z₂)^TWherein z is₁+z₂＝1，z_k≥0。

Constructing a unitary weight optimization model

Considering subjective factors which can be brought by the carrier members to the evaluation of the shippers, the subjective weight determination method is used for constructing an attribute weight optimization model by integrating big data mining technology, and the attribute weight optimization model comprises the following steps:

<math><mrow> <mi>s</mi> <mo>.</mo> <mi>t</mi> <mo>.</mo> <munderover> <mi>Σ</mi> <mrow> <mi>j</mi> <mo>=</mo> <mn>1</mn> </mrow> <mi>n</mi> </munderover> <msub> <mi>w</mi> <mi>j</mi> </msub> <mo>=</mo> <mn>1</mn> <mo>,</mo> <msub> <mi>w</mi> <mi>j</mi> </msub> <mo>&GreaterEqual;</mo> <mn>0</mn> <mo>,</mo> <mi>j</mi> <mo>&Element;</mo> <mi>N</mi> <mo>,</mo> </mrow></math>

where N is a set of attributes {1, 2.

The meaning of the model is to find a w_jSo that w_jAnd w_j ^kSum of squares of total partial variances L₁And minimum.

Construction of binary weight optimization model

In order to reduce human factors in the model as much as possible, multidimensional analysis is performed on data from the perspective of an objective weight determination method, and an attribute weight optimization model is constructed as follows:

G-minL＝(l₁,l₂,...l_m)

wherein,a_j ^*＝max{a_1j,a_2j,...a_mjis attribute P_jThe model can be simplified to be as follows by using an equal-weight linear weighting method for the ideal value of (1):

<math><mrow> <mi>s</mi> <mo>.</mo> <mi>t</mi> <mo>.</mo> <munderover> <mi>Σ</mi> <mrow> <mi>j</mi> <mo>=</mo> <mn>1</mn> </mrow> <mi>n</mi> </munderover> <msub> <mi>w</mi> <mi>j</mi> </msub> <mo>=</mo> <mn>1</mn> <msub> <mrow> <mo>,</mo> <mi>w</mi> </mrow> <mi>j</mi> </msub> <mo>&GreaterEqual;</mo> <mn>0</mn> <mo>,</mo> <mi>j</mi> <mo>&Element;</mo> <mi>N</mi> </mrow></math>

where M ═ 1, 2., M } is the set of preferred schemes, and N ═ 1, 2., N } is the set of attributes of the schemes.

Fourthly, synthesizing a single target optimization model

Integrating the two optimization models into G-min (L)₁,L₂) And converting the problem into a single-target optimization model as follows by a linear weighting method:

wherein, M ═ {1, 2., M } is the evaluation scheme set, and N ═ 1, 2., N } is the attribute set of the scheme. By constructing the lagrangian function, one can obtain:

wherein,

(4) solving for attribute weights

Due to a plurality of decision elements (T)_kK is 1,2) the value given to the overall weight of attribute j is:

is a schema property.

The weight vector of the attribute weight multivariate optimization model is as follows: w is a^*＝[w₁,w₂,...w_j,...,w_n]Wherein

The final weight matrix of the attribute set is A^*＝(A_ij)_m×n＝Bw^*T

Wherein A is_ijThe final weight value of the jth attribute of the credibility rating scheme i is B ═ B_ij)_m×nThe matrix is normalized for the set of attributes.

(5) Calculating the standard credibility score of the comparative evaluation scheme, and recording the time-credibility score curve of the transaction times

The final weight matrix is A according to the obtained attribute set^*＝(A_ij)_m×n＝Bw^*TAnd calculating a standard credibility score of the comparative scheme, and drawing a time-credibility score curve based on the transaction times. The main calculation steps are as follows:

calculating the credibility score of the credibility evaluation scheme i, wherein the formula is

Wherein, b_ijIs the attribute value a of the jth attribute of the scheme i_ijThe value after the normalization treatment is carried out,the weight of the j attribute of each scheme is obtained after a multi-element optimization model of the attribute weight is built.

② standardized evaluation scheme credibility score

The credibility evaluation scheme set comprises: (the credibility of the appraised goods owner, the credibility of the platform reference within the same industry), so the credibility of the appraisal scheme is divided into:

M＝(M₁,M₂,M₃)＝(M_{mean confidence level in the same industry}，M_{Credibility of goods evaluated owner}，M_{Platform reference confidence})

Wherein, 1 is equal to the equal industry average credibility, 2 is equal to the appraised goods owner credibility, and 3 is equal to the platform reference credibility

Confidence level M based on platform_{Platform reference confidence}For the standardized benchmark credibility score, the standardized credibility score m of each comparative scheme_i(i＝1,2,3，

1 is equal to the average credibility in the same industry, 2 is equal to the credibility of the goods owner to be evaluated, and 3 is equal to the standard credibility of the platform):

calculating credibility score of relative standard comparative scheme

Reference confidence m with standardized platform_{Platform reference confidence}The relative standard credibility score of each evaluation scheme is obtained by evaluating the standard credibility for the relative score

m_i ^*(i＝1,2,3，

m₁ ^*＝m₁-m₃

m₂ ^*＝m₂-m₃

m₃ ^*＝m₃-m₃＝0

m^*＝(m₁ ^*,m₂ ^*,m₃ ^*) Namely the final credibility scores of the evaluated comparative schemes.

Wherein, 1 is equal to the average credibility in the same industry, 2 is equal to the credibility of the goods owner to be evaluated, and 3 is equal to the benchmark credibility of the platform.

Fourthly, drawing a time-credibility score curve based on the transaction times

The credibility score is updated along with the updating of the transaction times of the platform owner, and the platform reference credibility score is 0 and remains unchanged according to the calculation result, so that the method has practical significance; the credibility of the same industry can be continuously updated along with the accumulation of transaction data of different industries (or the whole platform); the credibility of the evaluated goods owner changes along with the increase of the transaction times of the goods owner on the platform. Therefore, by taking time as a horizontal axis and taking the credibility score as a vertical axis, a time-credibility score curve based on transaction times can be obtained, the credibility condition of the owners of goods in the same industry (or the whole platform) in a period of time and the credibility change condition of a certain owner of goods can be intuitively obtained according to the curve, and the platform members can be helped to better control risks and select the owners of goods to receive orders.

(6) Calculating harmony indexes among the more evaluated schemes and relative scores of the attributes

The platform provides more diversified and humanized choices for different platform members by calculating the harmony index and the relative standardized scores of the attributes in order to meet the risk preferences or selection preferences of the different platform members. The main calculation steps are as follows:

calculating a harmony index

Harmony index J_ikThe evaluation method refers to the proportion of the sum of the attribute weights of the evaluation scheme i not inferior to the scheme k in the sum of all weights, and if the scheme i is superior to the scheme k (i is 1, 2.. the sum of m, k ≠ k) according to the attribute j (j is 1, 2.. the sum of all weights), the set of all the attributes j meeting i > k is recorded as U > k, and therefore the set of all the attributes j meeting i > k is recorded as U > k⁺(i, k) can also give U^＝(i, j), as follows:

U⁺(i,k)＝{j\1≤j≤n,A_ij＞A_ik}

U^＝(i,k)＝{j\1≤j≤n,A_ij＝A_ik}

the harmonicity index is then formulated as:

<math><mrow> <msub> <mi>J</mi> <mi>ik</mi> </msub> <mo>=</mo> <mfrac> <mrow> <munder> <mi>Σ</mi> <mrow> <mi>j</mi> <mo>&Element;</mo> <msup> <mi>U</mi> <mo>+</mo> </msup> <mrow> <mo>(</mo> <mi>i</mi> <mo>,</mo> <mi>k</mi> <mo>)</mo> </mrow> </mrow> </munder> <msub> <mi>A</mi> <mi>ij</mi> </msub> <mo>+</mo> <munder> <mi>Σ</mi> <mrow> <mi>j</mi> <mo>&Element;</mo> <msup> <mi>U</mi> <mo>=</mo> </msup> <mrow> <mo>(</mo> <mi>i</mi> <mo>,</mo> <mi>k</mi> <mo>)</mo> </mrow> </mrow> </munder> <msub> <mi>A</mi> <mi>ij</mi> </msub> </mrow> <mrow> <munderover> <mi>Σ</mi> <mrow> <mi>j</mi> <mo>=</mo> <mn>1</mn> </mrow> <mi>n</mi> </munderover> <msub> <mi>A</mi> <mi>ij</mi> </msub> </mrow> </mfrac> </mrow></math>

calculating relative scores of each attribute of the owner

The final weight matrix is A according to the obtained attribute set^*＝(A_ij)_m×n＝Bw^*TThe relative normalized score of each attribute of the owner is a_2j ^*＝A_2j-A_3jWherein j is 1,2

Finally, the relative standard of each attribute of the owner is obtained to obtain the diversity a₂ ^*＝(a₂₁ ^*,a₂₂ ^*,...a_2n ^*)。

Has the advantages that: compared with the prior art, the method is suitable for evaluating the credibility of the goods owner in the electronic commerce platform for providing the logistics transaction service. The method is characterized in that a plurality of attributes influencing the reliability of a shipper are classified into mutually-connected ordered levels to be organized, the platform reference reliability is set according to the mass data of the carrier members and the platform, the average reliability of a certain industry (or the whole platform) and the reliability of a certain shipper are compared with each other to obtain a relative standardized reliability score, and the result is more visual. The credibility attribute of the goods owner and the platform benchmark attribute score are compared pairwise, and then each attribute is quantitatively described, so that the platform member can select and grasp risks better. According to the evaluation result, the comprehensive credibility condition of the goods owner can be objectively obtained, and the problem of some information asymmetry between the customer and the merchant is solved to a certain extent, so that platform members or other members can quickly and objectively find merchants with high credit among countless merchants and make selections meeting the needs of the platform members or other members.

Drawings

FIG. 1 is a flow chart of a method of the present invention;

FIG. 2 is a review scenario and attribute model for an embodiment of the present invention;

FIG. 3 is a time-confidence score curve based on transaction times.

Detailed Description

The present invention is further illustrated by the following examples, which are intended to be purely exemplary and are not intended to limit the scope of the invention, as various equivalent modifications of the invention will occur to those skilled in the art upon reading the present disclosure and fall within the scope of the appended claims.

And selecting credibility evaluation scheme attributes required by the platform by referring to the database and the industry indexes, and establishing a scheme set and an attribute set according to the credibility evaluation scheme and the mutual relation among the attributes, as shown in figure 2. Credibility evaluation scheme set: the credibility of the same industry, the credibility of the goods to be evaluated and the credibility of the platform reference. Attribute set: seven evaluation indexes, namely punctual shipment of a shipper, punctual discharge of a consignee, the number of times of a shipper service, the number of times of complaints of members of a subject platform, the information release timeliness rate, the payment timeliness rate of the shipper payment of the shipper and the one-time quoted price and order rate, are selected as attributes influencing the scheme. The evaluation schemes and the evaluation conditions of the attributes of the present example are shown in Table 1.

Table 1 evaluation table of credibility comparison scheme attributes in the embodiment

(2) Attribute set normalization process

Cost type attribute processing: b_ij＝(a_j ^max-a_ij)/(a_j ^max-a_j ^min)

The credibility comparison scheme attribute decision matrix is obtained according to table 1 as follows:

the number of times of carrying business by a shipper, the information release timeliness rate, the shipper payment timeliness rate and the one-time quotation order-forming rate are benefit attributes, and the larger the evaluation value is, the better the evaluation value is; the consignor on-time loading, the consignee on-time unloading and the complaint times of the members of the platform are cost attributes, and the smaller the evaluation value, the better. After the matrix A is subjected to standardization processing, the obtained standardized matrix is as follows:

Determining multivariate weights

In order to reduce the influence of single preference and cognitive limitation on results, the platform carries out deep mining and analysis on massive big data evaluated by the carrier members on the shippers to obtain the most intuitive comprehensive evaluation of the carrier members on the shippers; in order to reduce the deviation caused by evaluation of a single dimension, a platform decision element is introduced from another dimension, and the credibility of a shipper is subjected to supplementary evaluation based on the historical trading data and trading behaviors of the platform. And sets reasonable importance coefficients for the two decision elements based on the completeness and authenticity of the data.

In this example, the carrier member and the platform are weighted and averaged for each attribute, and then normalized to obtain:

the carrier member decision element is assigned with the attribute weight as follows: w is a¹＝(0.53,0.13,0.1,0.6,0.1,0.64,0.64)^T

The attribute weight assigned by the platform decision element is as follows: w is a²＝(0.2,0.2,0.4,0.75,0.15,0.9,0.4)^T

The importance degree of each decision element is as follows: z ═ 0.45,0.55)^T。

(4) Solving for attribute weights

In order to make the attribute weight include both subjective preference and objective information, the subjective weight optimization model and the objective weight optimization model may be integrated into G ═ min (L)₁,L₂) And converting the problem into a single-target optimization model as follows by a linear weighting method:

<math><mrow> <mi>s</mi> <mo>.</mo> <mi>t</mi> <mo>.</mo> <munderover> <mi>Σ</mi> <mrow> <mi>j</mi> <mo>=</mo> <mn>1</mn> </mrow> <mi>n</mi> </munderover> <msub> <mi>w</mi> <mi>j</mi> </msub> <mo>=</mo> <mn>1</mn> <msub> <mrow> <mo>,</mo> <mi>w</mi> </mrow> <mi>j</mi> </msub> <mo>&GreaterEqual;</mo> <mn>0</mn> <mo>,</mo> <mi>i</mi> <mo>&Element;</mo> <mi>M</mi> <mo>,</mo> <mi>j</mi> <mo>&Element;</mo> <mi>N</mi> </mrow></math>

wherein, M ═ {1, 2., M } is the evaluation scheme set, and N ═ 1, 2., N } is the attribute set of the scheme.

By constructing the lagrangian function, one can obtain:

c＝(0.89,0.89,0.33,0.67,0.4,0.22,0.1)^T

d＝(0.15,0.1,0.11,0.31,0.06,0.38,0.25)^T

according to the formula, the weight vector which reflects the information of the carrier member and the platform at the same time can be obtained as follows:

w^*＝(0.23,0.19,0.07,0.28,0.07,0.11,0.04)^T

calculating the credibility score of the credibility comparison scheme i by the formula

Solving the following formula:

M＝(M₁,M₂,M₃)＝(M_{mean confidence level in the same industry}，M_{Credibility of goods evaluated owner}，M_{Platform reference confidence})＝(0.5,0.75,0.28)

Credibility score of standardized comparatively assessed scheme

thus, m is obtained as (1.76,2.64,1)

Calculating a relative standardized comparatively assessed scheme credibility score

m_i ^*(i＝1,2,3，

m₁ ^*＝m₁-m₃

m₂ ^*＝m₂-m₃

m₃ ^*＝m₃-m₃＝0

to obtain m^*＝(m₁ ^*,m₂ ^*,m₃ ^*) The final credibility score of each evaluated scheme is (0.76,1.64, 0). Wherein, 1 is equal to the average credibility in the same industry, 2 is equal to the credibility of the goods owner to be evaluated, and 3 is equal to the benchmark credibility of the platform.

As shown in FIG. 3, a time-confidence score curve based on transaction times is plotted

(6) Calculating relative scores for attributes of a owner

The final weight matrix is A according to the obtained attribute set^*＝(A_ij)_m×n＝Bw^*TThe relative normalized score of each attribute of the owner is a_2j ^*＝A_2j-A_3jWherein, j is 1,2, 7, 1 is the punctual shipment of the shipper, 2 is the punctual discharge of the consignee, 3 is the number of shipper delivery business times, 4 is the number of complaints of the platform member, 5 is the information distribution timeliness rate, 6 is the payment timeliness rate of the shipper's payment, and 7 is the one-time quotation invoicing rate.

And finally obtaining the relative standard scores of each attribute of the owner:

a₂ ^*＝(a₂₁ ^*,a₂₂ ^*,...a_2n ^*)＝(0.2,0.1,0.04,0,0.07,0,0.04)

wherein, a₂₄ ^*＝0，a₂₆ ^*The two attribute scores of 0 indicating the 'complaint times of platform members' of the evaluated owner and 'payment timeliness rate of owner' are the same as the platform benchmark score, and the 0 point here indicates that the owner just passes the good in the attribute relative to the benchmark level, and the platform member should pay attention to the high credibility risk of the owner in the two aspects and be careful when selecting.

Claims

1. A comprehensive and multidimensional owner selection and quantification method is characterized by comprising the following steps:

(2) Attribute set normalization process

Carrying out non-dimensionalization on each attribute of the scheme, wherein the non-dimensionalization of the benefit attribute and the cost attribute is as follows:

Cost type attribute processing: b_ij＝(a_j ^max-a_ij)/(a_j ^max-a_j ^min)

Wherein, a_ijIs the attribute value of the jth attribute of the scheme i, b_ijIs a_ijNormalized value of a_j ^maxIs the jth attribute P_jMaximum value of a_j ^minIs P_jIs measured. b_ijE (0,1), i ═ 1, 2., m, j ═ 1, 2., n, normalized matrix B ═ B_ij)_m×n；

Introducing a multivariate optimization model, comprehensively considering different influences of the multivariate optimization model on the credibility of the shipper from two different dimensions of the carrier member and the platform, and carrying out mathematical description on the multivariate optimization model on the shipper member and the platform on the basis of big data analysis; the contact points of the carrier members and the platform with the shipper are different, so that different attribute evaluation values are shown for the emphasis points of the reliability of the shipper;

(4) solving for attribute weights

The final weight matrix of the attribute set is A^*＝(A_ij)_m×n＝Bw^*T

Wherein A is_ijThe final weight value of the jth attribute of the credibility rating scheme i is B ═ B_ij)_m×nNormalizing the matrix for the attribute set;

The final weight matrix is A according to the obtained attribute set^*＝(A_ij)_m×n＝Bw^*TCalculate the comparative commentThe scheme standardizes the credibility score and draws a time-credibility score curve based on the transaction times;

The platform provides more diversified and humanized choices for different platform members by calculating the harmony index and the relative standardized scores of the attributes in order to meet the risk preferences or selection preferences of the different platform members.

2. The comprehensive, multi-dimensional owner selection quantification method of claim 1, wherein the specific steps of step 3 are:

(ii) multivariate weight determination

In order to reduce the influence of single preference and cognitive limitation on results, the platform carries out weighted average on historical evaluation scores of all the carrier members on each index of the shipper to obtain the most intuitive comprehensive evaluation score of the carrier members on the shipper; in order to reduce the deviation caused by single-dimension evaluation, a platform decision element is introduced from another dimension, and the credibility of a buyer is subjected to supplementary evaluation and scoring based on the historical trading data and the trading behavior of the platform; then, carrying out standardization processing on the scores to obtain the weight value of each attribute, and finally setting importance degree coefficients for the two decision elements based on the completeness and the authenticity of data;

decision element T determined by two dimensions_kK is 1, 2; wherein T is₁Representative Carrier Member decision element, T₂A representative platform decision element;

The importance degree of each decision element is as follows: z is (z)₁,z₂)^TWherein z is₁+z₂＝1，z_k≥0；

Constructing a unitary weight optimization model

<math> <mfenced open='' close=''> <mtable> <mtr> <mtd> <mi>s</mi> <mo>.</mo> <mi>t</mi> <mo>.</mo> </mtd> <mtd> <munderover> <mi>Σ</mi> <mrow> <mi>j</mi> <mo>=</mo> <mn>1</mn> </mrow> <mi>n</mi> </munderover> <msub> <mi>w</mi> <mi>j</mi> </msub> <mo>=</mo> <mn>1</mn> </mtd> <mtd> <msub> <mi>w</mi> <mi>j</mi> </msub> <mo>&GreaterEqual;</mo> <mn>0</mn> <mo>,</mo> <mi>j</mi> <mo>&Element;</mo> <mi>N</mi> </mtd> </mtr> </mtable> </mfenced> </math>

where N is the attribute set of the scheme {1,2

The meaning of the model is to find onew_jSo that w_jAnd w_j ^kSum of squares of total partial variances L₁Minimum;

construction of binary weight optimization model

From the perspective of an objective weight determination method, multidimensional analysis is performed on data, and an attribute weight optimization model is constructed as follows:

G-minL＝(l₁,l₂,...l_m)

wherein,

is attribute P_jThe model can be simplified to be as follows by using an equal-weight linear weighting method for the ideal value of (1):

<math> <mfenced open='' close=''> <mtable> <mtr> <mtd> <mi>s</mi> <mo>.</mo> <mi>t</mi> <mo>.</mo> </mtd> <mtd> <munderover> <mi>Σ</mi> <mrow> <mi>j</mi> <mo>=</mo> <mn>1</mn> </mrow> <mi>n</mi> </munderover> <msub> <mi>w</mi> <mi>j</mi> </msub> <mo>=</mo> <mn>1</mn> </mtd> <mtd> <msub> <mi>w</mi> <mi>j</mi> </msub> <mo>&GreaterEqual;</mo> <mn>0</mn> <mo>,</mo> <mi>i</mi> <mo>&Element;</mo> <mi>M</mi> <mo>,</mo> <mi>j</mi> <mo>&Element;</mo> <mi>N</mi> </mtd> </mtr> </mtable> </mfenced> </math>

where M is a set of comparatively rated schemes, and N is a set of attributes of the schemes, where M is a set of comparatively rated schemes and N is a set of attributes of the schemes

Fourthly, synthesizing a single target optimization model

wherein,

3. a comprehensive, multi-dimensional owner selection quantification method according to claim 2, wherein the main calculation steps of step (5) are as follows:

Wherein, b_ijIs the attribute value a of the jth attribute of the scheme i_ijThe value after the normalization treatment is carried out,the weight of the j attribute of each scheme is obtained after a multi-element optimization model of the attribute weight is constructed;

② standardized evaluation scheme credibility score

calculating credibility score of relative standard comparative scheme

m_i ^*(i＝1,2,3，

m₁ ^*＝m₁-m₃

m₂ ^*＝m₂-m₃

m₃ ^*＝m₃-m₃＝0

m^*＝(m₁ ^*,m₂ ^*,m₃ ^*) The final credibility scores of the evaluated comparative schemes are obtained;

wherein, 1 is the equal credibility in the same industry, 2 is the credibility of the evaluated goods owner, and 3 is the standard credibility of the platform;

fourthly, drawing a time-credibility score curve based on the transaction times

4. A comprehensive, multi-dimensional owner selection quantification method as claimed in claim 1, wherein the step (6) mainly comprises the following steps:

calculating a harmony index

Harmony index J_ikThe evaluation scheme i is not inferior to the proportion of the sum of the attribute weights of the scheme k in the sum of all weights, and is marked as being superior to the scheme k (i 1, 2..., m, k 1, 2..., m, i ≠ k) according to the attribute j (j 1, 2..., n)Thereby satisfying allIs recorded as U⁺(i, k) can also give U^＝(i, k), specifically as follows:

the harmonicity index is then formulated as:

<math> <mrow> <msub> <mi>J</mi> <mi>ik</mi> </msub> <mo>=</mo> <mfrac> <mrow> <munder> <mi>Σ</mi> <mrow> <mi>j</mi> <mo>&Element;</mo> <msup> <mi>U</mi> <mo>+</mo> </msup> <mrow> <mo>(</mo> <mi>i</mi> <mo>,</mo> <mi>k</mi> <mo>)</mo> </mrow> </mrow> </munder> <msub> <mi>A</mi> <mi>ij</mi> </msub> <mo>+</mo> <munder> <mi>Σ</mi> <mrow> <mi>j</mi> <mo>&Element;</mo> <msup> <mi>U</mi> <mo>=</mo> </msup> <mrow> <mo>(</mo> <mi>imk</mi> <mo>)</mo> </mrow> </mrow> </munder> <msub> <mi>A</mi> <mi>ij</mi> </msub> </mrow> <mrow> <munderover> <mi>Σ</mi> <mrow> <mi>i</mi> <mo>=</mo> <mn>1</mn> </mrow> <mi>n</mi> </munderover> <msub> <mi>A</mi> <mi>ij</mi> </msub> </mrow> </mfrac> </mrow> </math>

calculating relative scores of each attribute of the owner

The final weight matrix is A according to the obtained attribute set^*＝(A_ij)_m×n＝Bw^*TThe relative normalized score of each attribute of the owner is a_2j ^*＝A_2j-A_3jWherein j is 1, 2.. and n is a scheme attribute;