CN107730080A - A kind of third party's food inspection mechanism evaluation method for government buying service - Google Patents
A kind of third party's food inspection mechanism evaluation method for government buying service Download PDFInfo
- Publication number
- CN107730080A CN107730080A CN201710835219.7A CN201710835219A CN107730080A CN 107730080 A CN107730080 A CN 107730080A CN 201710835219 A CN201710835219 A CN 201710835219A CN 107730080 A CN107730080 A CN 107730080A
- Authority
- CN
- China
- Prior art keywords
- evaluation
- party
- data set
- food inspection
- inspection mechanism
- 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.)
- Pending
Links
- 238000011156 evaluation Methods 0.000 title claims abstract description 89
- 230000007246 mechanism Effects 0.000 title claims abstract description 48
- 238000007689 inspection Methods 0.000 title claims abstract description 44
- 238000004364 calculation method Methods 0.000 claims abstract description 6
- 238000012360 testing method Methods 0.000 claims description 18
- 239000011159 matrix material Substances 0.000 claims description 16
- 238000004458 analytical method Methods 0.000 claims description 13
- 238000012545 processing Methods 0.000 claims description 5
- 238000010276 construction Methods 0.000 claims description 4
- 238000013507 mapping Methods 0.000 claims description 4
- 238000005192 partition Methods 0.000 claims description 4
- 238000013210 evaluation model Methods 0.000 claims description 3
- 238000000034 method Methods 0.000 abstract description 14
- 230000009286 beneficial effect Effects 0.000 abstract description 3
- 230000009897 systematic effect Effects 0.000 abstract 1
- 238000001514 detection method Methods 0.000 description 5
- 238000005516 engineering process Methods 0.000 description 5
- 238000007726 management method Methods 0.000 description 4
- 241000422846 Sequoiadendron giganteum Species 0.000 description 2
- 235000021393 food security Nutrition 0.000 description 2
- 230000008569 process Effects 0.000 description 2
- 230000004888 barrier function Effects 0.000 description 1
- 230000008901 benefit Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000007621 cluster analysis Methods 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 230000002860 competitive effect Effects 0.000 description 1
- 238000007405 data analysis Methods 0.000 description 1
- 238000013480 data collection Methods 0.000 description 1
- 230000007812 deficiency Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 230000008520 organization Effects 0.000 description 1
- 238000004445 quantitative analysis Methods 0.000 description 1
- 239000002994 raw material Substances 0.000 description 1
- 238000012552 review Methods 0.000 description 1
- 210000005182 tip of the tongue Anatomy 0.000 description 1
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
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0639—Performance analysis of employees; Performance analysis of enterprise or organisation operations
- G06Q10/06393—Score-carding, benchmarking or key performance indicator [KPI] analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
- G06F18/231—Hierarchical techniques, i.e. dividing or merging pattern sets so as to obtain a dendrogram
-
- 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
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/10—Services
- G06Q50/26—Government or public services
Landscapes
- Business, Economics & Management (AREA)
- Engineering & Computer Science (AREA)
- Human Resources & Organizations (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Tourism & Hospitality (AREA)
- Strategic Management (AREA)
- Economics (AREA)
- Educational Administration (AREA)
- Development Economics (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Marketing (AREA)
- Entrepreneurship & Innovation (AREA)
- General Business, Economics & Management (AREA)
- Bioinformatics & Computational Biology (AREA)
- Operations Research (AREA)
- General Engineering & Computer Science (AREA)
- Evolutionary Biology (AREA)
- Game Theory and Decision Science (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Artificial Intelligence (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Quality & Reliability (AREA)
- Life Sciences & Earth Sciences (AREA)
- Evolutionary Computation (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Primary Health Care (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The present invention relates to the assessment technique field to food inspection mechanism, specifically disclose a kind of third party's food inspection mechanism evaluation method for government buying service, analyzed by introducing rough set theory, with reference to the clustering of Kruskal maximal trees, and from the weighted value of each evaluation index of entropy angle calculation, the authority of expert opinion and the objectivity of user's evaluation are made full use of simultaneously, realize the objective evaluation to third party's food inspection mechanism of government buying service.Compared with prior art, the present invention can carry out systematic objective evaluation to third party's food inspection mechanism, and beneficial reference and reference are provided for government buying service.
Description
Technical field
The present invention relates to the assessment technique field to food inspection mechanism, it is particularly a kind of for government buying service the
Tripartite's food inspection mechanism evaluation method.
Background technology
The society of food-safety problem controls altogether has reached widespread consensus in the Chinese government and educational circles, by transferring society
Resource, which participates in food safety Regulation, can overcome resource constraint, reduce executive cost, improve supervisory efficiency.Government buying service is
Social supervision resource is introduced, promotes the important way of food security governance.So-called government buying service refers to that government passes through
The public service that script is undertaken by itself is handed to social organization, enterprises and institutions' list by the forms such as competitive bidding, orientation commission, invitation to bid
Position is fulfiled, and to improve the service efficiency of the quality of Public Services Supply and financial fund, is improved governance structure, is met the public
Diversification, individual demand.For food safety Regulation, strengthen government buying service processes and be a timely one, attract more
The society that more societal forces are added to food security controls general layout altogether, can effectively overcome current supervision problem, ensures wide
Big people's " safety on the tip of the tongue ".
Third party's food inspection mechanism refers to by the associated mechanisms specified of country as independent third party, to food production from
Raw material carries out notarial survey to finished product links, it is ensured that a kind of systems of product quality.It has been widely used at present
In the food safety Regulation of government buying service.However, third party's food inspection market in China is just started to walk, with foreign countries
Mature market is compared, and many problems and obstruction also be present：
1) quite a few food inspection laboratory of China appears in market with quasi official posture always.Due to being subordinated to
The public institution that different government departments divides into, administrative and financially all relevant, mechanism is difficult to accomplish science in the market
Independence.
2) China third party testing agency is at present also in disorderly competition state, though national departments concerned has started to attention
Tripartite's detection pattern and this block market, but one cuts and is in the starting stage, these emerging testing agencies are in default of effective
Supervision, put forth effort realization profit of opening up markets again, brand awareness is not strong, and market is not standardized also.
3) the inspection testing agency that China presently, there are is still in the early stage of development, and small scale, layout dissipate, mechanism repeats to build
If practitioner's quality is uneven, detection level differs.Barrier between different departments is obvious, and information communication is not smooth, assists government regulation
Consciousness is weaker, and internationalization level is not high, the simple repetition of a large amount of low-level resources.
At present, State Council has issued《Establishment of social credit system planning outline (2014-2020)》, clearly propose：
" emphasis is strengthened examining the credit classification management of detection Lei Deng mechanisms, and scientific and reasonable evaluation index system, appraisal system are established in exploration
And working mechanism ".Food inspection is the important component of modern service industry, is the important technical for ensureing food quality.
If food inspection mechanism makes the behavior for violating the principle of good faith, the reputation of mechanism can be not only influenceed, serious can even endanger
To the existence of mechanism.
However, because the factor for influenceing food inspection mechanism is a lot, and much influence factor is unclear, is not easy to quantify, it is existing
Food inspection mechanism evaluation be generally divided into including expert opinion, peer review, user evaluation.Wherein user's (i.e. government buying
Service side) as the direct contact object of testing agency and service object, service quality and service level to testing agency are most
Has right to speak；And expert opinion then has preferably authority.At present, third party's food inspection mechanism of government buying service is commented
Valency mainly in a manner of subjective assessment based on, method for objectively evaluating is less.And objective evaluation carries out weight due to adopting objective data
Analysis, there is higher confidence level, and the main way of current detection mechanism evaluation.
《Examine the discussion of testing agency's Credit Evaluation System model》(quality technical supervision is studied, 2016,3 (45)：58-60) carry
Go out to use analytic hierarchy process (AHP) (AHP), determine to influence to examine testing agency sincere in terms of law, technology, management, responsibility etc. four
Weight shared by each key element, it is evaluated as examining the sincere value of testing agency to carry out quantitative analysis with reference to each key element, provides sincerity and comment
Point, but the acquisition of weight still relies on the subjective opinion of expert, and it is unreasonable and cause evaluation conclusion to easily cause evaluation criterion weight
There is larger difference.
The content of the invention
The invention aims to solve the deficiency of prior art problem, there is provided it is a kind of for government buying service the 3rd
Square food inspection mechanism evaluation method.
To reach above-mentioned purpose, the present invention is implemented according to following technical scheme：
A kind of third party's food inspection mechanism evaluation method for government buying service, comprise the following steps：
Step 1: gather first to multiple expert opinion history data sets for being evaluated third party's food inspection mechanism
User's evaluation history data set S1, test data set S to be evaluated2；
Step 2: pretreatment expert opinion history data set
Step 3: calculate expert opinion history data set after pretreatmentSimilarity relation matrix, generate Kruskal
Maximal tree, obtain the clustering result of set of data objects；
Step 4: according to data object clustering result and user's evaluation data set S of Corresponding matching1Construction evaluation is determined
Plan table S；
Step 5: ask for each user's evaluation index a ∈ C importance SGF (a, A, D) and each evaluation index rough set believe
Cease the weight metric value w under entropya；
Step 6: it is based on test data set S to be evaluated2Summation evaluation analysis is weighted with each evaluation criterion weight, is obtained
To object x to be evaluatedl'∈S2Decision-Making Evaluation valueWherein wa(h)For the weight of h-th of evaluation index,
xlhFor value of l-th of object in h-th of evaluation index.
Wherein, to expert opinion history data set in the step 2Data prediction include：For profit evaluation model category
Property, property value according toProcessing；For cost type attribute, then according toProcessing.
Wherein, comprised the following steps that in the step 3：
1) formula is utilizedCalculate expert opinion history data set after pretreatment
Similarity relation matrix R, wherein, R=(rij)n×n, c suitably chooses so that 0≤rij≤1；
2) the maximum r of value is found in similarity relation matrix R non-leading diagonalijAnd corresponding data object ri、rj,
And drawLocal Kruskal maximal trees；
3) repeat step 2), it is connected to all summits, and without circle, so as to obtain overall Kruskal most
Big tree；
4) it is based on to pretreated expert opinion history data setIn property value setting threshold value λ ∈ [0,1], cut
Connection weight r in disconnected maximal treeijBranch less than λ, a disconnected graph is obtained, and each connected component is just constituted in λ-level
Clustering result；
5) data object under same clustering is given to the mark of same integer numerical value, expert opinion is completed and goes through
History data setDivision.
Wherein, in the step 4, drawn to obtaining each cluster for being evaluated third party's food inspection mechanism in step 3
Fractional value value, as the decision attribute D in decision table S and corresponding value, then gather multiple users i.e. government procurement and take
Business side is conditional attribute C to expert opinion history data set by given indexIncluded in third party's food inspection mechanism enter
Capable marking, i.e. user's evaluation history data set S1, by user's evaluation history data set S1In every a data object composition it is special
Family's evaluation history data setWhat is obtained during cluster belongs to the decision attribute values of each evaluation object, structure evaluation decision table S=
(U, R, V, f) is a decision table, wherein, U is entire objects set；R=C ∪ D are conditional attribute collection C's and decision attribute D
Union；V is property set R codomain, i.e.,f:U × R --- → V is referred to as mapping function, i.e.,There is f
(xi, a)=Va。
Wherein, the step 5 comprises the following steps：
1) set X=U/IND (Q)={ X is made1,X2,...,XpAnd Y=U/IND (P)={ Y1,Y2,...,YqBe respectively
Object set U derived equivalent partitions in the case where knowledge is property set P and Q, the probability distribution for defining its dividing subset are p (Xp)=| Xp
|/|U|,p(Xq)=| Yq|/| U |, wherein | | for the gesture of set；
2) conditional information entropy under knowledge Q divides relative to knowledge P is definedWherein, p (Yt/Xs)=| Yt∩Xs|/|Xs|；
3) basis based on more than, obtains the calculation formula of conditional attribute importance：WhereinThen should
Conditional attribute a importance measures are SGF (a, A, D)=H (D/A)-H (D/A ∪ { a }), and then obtain conditional attribute a thick
Weight metric value under rough collection comentropy is：
Compared with prior art, the present invention is analyzed by introducing rough set theory, with reference to the poly- of Kruskal maximal trees
Class divides, and from the weighted value of each evaluation index of entropy angle calculation, while make full use of the authority of expert opinion and user to comment
The objectivity of valency, the objective evaluation to third party's food inspection mechanism of government buying service is realized, carried for government buying service
It is provided with the reference and reference of benefit.
Brief description of the drawings
Fig. 1 is the rough set evaluation analysis flow chart of the present invention.
Fig. 2 is the Kruskal maximum tree graphs of experimental analysis in embodiment of the present invention.
Embodiment
With reference to specific embodiment, the invention will be further described, the illustrative examples and explanation invented herein
For explaining the present invention, but it is not as a limitation of the invention.
As shown in figure 1, a kind of third party's food inspection mechanism evaluation method for government buying service of the present invention, bag
Include following steps：
First, rough set is asked for
(1) decision attribute obtains
1st, primary data matrix：Remember that data object entirety is T, xi∈ T (i=1,2 ..., n), if each object to be clustered
xiAll portrayed by m Criterion Attribute, i.e. xi=(xi1,xi2,...,xim), then n × m ties up matrix X- (xih)n×m(h=1,
2 ..., m) it is referred to as primary data matrix.
2nd, similarity relation matrix：Remember R=(rij)n×nFor the similarity relation matrix on domain T, wherein,
C suitably chooses so that 0≤rij≤1。
3rd, the clustering based on Kruskal maximal trees is realized：
Input：Set of data objects T to be clustered, threshold value λ；
Output：The clustering result of set of data objects.
Realize that step is as follows：
Step1：Data-oriented object set T, similarity relation matrix R is constructed, in similarity relation matrix R non-leading diagonal
Find the maximum r of valueijAnd corresponding data object ri、rj, and it is as followsDraw local Kruskal maximal trees.
Step2：Repeat step step1, it is connected to all summits, and without circle, it is overall so as to obtain
Kruskal maximal trees.
Step3：Threshold value λ ∈ [0,1] based on setting, break apart by chopping connection weight r in maximal treeijBranch less than λ, obtain one
Disconnected graph, and each connected component just constitutes the clustering result in λ-level.
Step4：Data object under same clustering is given to the mark of same integer numerical value, completes data pair
Division as collecting T.
Therefore, the mode of the clustering based on Kruskal maximal trees, it is possible to achieve the decision attribute of data object is really
It is fixed, and its decision attribute values is used as using the integer numeric indicia.
(2) attribute weight obtains
Operations in Rough Set Theory intuitive under information view is stronger, and the information definition of its Importance of Attributes contains its generation
Number definition, therefore the present invention gives objective ask from the angle of comentropy based on rough set theory to the weight of each conditional attribute
Take.
1st, decision table：Remember that S=(U, R, V, f) is a decision table.Wherein, U is entire objects set；R=C ∪ D are bar
Part property set C and decision attribute D union；V is property set R codomain, i.e.,f:U × R --- → V is referred to as mapping
Function, i.e.,
There are f (xi, a)=Va。
2nd, probability distribution：If set X=U/IND (Q)={ X1,X2,...,XpAnd Y=U/IND (P)={ Y1,Y2,...,
YqIt is respectively that object set U derived equivalent partition, the probability distribution for defining its dividing subset under knowledge (property set) P and Q are
p(Xp)=| Xp|/|U|,p(Xq)=| Yq|/| U |, wherein | | for the gesture of set.
3rd, conditional information entropy：Define the conditional information entropy under knowledge Q divides relative to knowledge P
Wherein, p (Yt/Xs)=| Yt∩Xs|/|Xs|。
The basis based on more than, provide the calculation formula of conditional attribute importance：WhereinThen this
Part attribute a importance measures are SGF (a, A, D)=H (D/A)-H (D/A ∪ { a }).It is possible to further obtain the condition category
Weight metric values of the property a under rough sets for information entropy be：
With reference to above-mentioned theory, the present invention will be described in detail, and flow chart is as shown in figure 1, comprise the following steps that：
Step 1: gather first to multiple expert opinion history data sets for being evaluated third party's food inspection mechanism
User's evaluation history data set S1, test data set S to be evaluated2；
Step 2: pretreatment expert opinion history data setTo obtain the decision-making category of the decision table in Rough Set Analysis
Property, it should first gather expert opinion history data setBut due to different expert estimations be accustomed to it is different, often had to score value compared with
Big difference, easily cause expert opinion importance difference and change the physical meaning of primary data, to expert opinion history data set
Data prediction include：For profit evaluation model attribute, property value according toProcessing；For into
This type attribute, then according toProcessing；
Step 3: calculate expert opinion history data set after pretreatmentSimilarity relation matrix, generate Kruskal
Maximal tree, the clustering result of set of data objects is obtained, is concretely comprised the following steps：
1) expert opinion history data set after pretreatment is calculated using formula (1)Similarity relation matrix R, wherein,
R=(rij)n×n, c suitably chooses so that 0≤rij≤1；
2) the maximum r of value is found in similarity relation matrix R non-leading diagonalijAnd corresponding data object ri、rj,
And drawLocal Kruskal maximal trees；
3) repeat step 2), it is connected to all summits, and without circle, so as to obtain overall Kruskal most
Big tree；
4) it is based on to pretreated expert opinion history data setIn property value setting threshold value λ ∈ [0,1], cut
Connection weight r in disconnected maximal treeijBranch less than λ, a disconnected graph is obtained, and each connected component is just constituted in λ-level
Clustering result；
5) data object under same clustering is given to the mark of same integer numerical value, expert opinion is completed and goes through
History data setDivision；
Step 4: according to data object clustering result and user's evaluation data set S of Corresponding matching1Construction evaluation is determined
Plan table S, it is specially：To obtaining each clustering numerical value value for being evaluated third party's food inspection mechanism in step 3, make
For the decision attribute D in decision table S and corresponding value, multiple users i.e. government procurement service side is then gathered by given index
I.e. conditional attribute C is to expert opinion history data setIncluded in the marking that carries out of third party's food inspection mechanism, i.e. user
Evaluation history data set S1, by user's evaluation history data set S1In every a data object composition expert opinion historical data
CollectionWhat is obtained during cluster belongs to the decision attribute values of each evaluation object, and structure evaluation decision table S=(U, R, V, f) is one
Individual decision table, wherein, U is entire objects set；R=C ∪ D are conditional attribute collection C and decision attribute D union；V is property set
R codomain, i.e.,f:U × R --- → V is referred to as mapping function, i.e.,There are f (xi, a)=Va；
Step 5: ask for each user's evaluation index a ∈ C importance SGF (a, A, D) and each evaluation index rough set believe
Cease the weight metric value w under entropya, concretely comprise the following steps：
1) set X=U/IND (Q)={ X is made1,X2,...,XpAnd Y=U/IND (P)={ Y1,Y2,...,YqBe respectively
Object set U derived equivalent partitions in the case where knowledge is property set P and Q, the probability distribution for defining its dividing subset are p (Xp)=| Xp
|/|U|,p(Xq)=| Yq|/| U |, wherein | | for the gesture of set；
2) conditional information entropy under knowledge Q divides relative to knowledge P is definedWherein, p (Yt/Xs)=| Yt∩Xs|/|Xs|；
3) basis based on more than, obtains the calculation formula of conditional attribute importance：WhereinThen should
Conditional attribute a importance measures are SGF (a, A, D)=H (D/A)-H (D/A ∪ { a }), and then obtain conditional attribute a thick
Weight metric value under rough collection comentropy is：
Step 6: it is based on test data set S to be evaluated2Summation evaluation analysis is weighted with each evaluation criterion weight, is obtained
To object x to be evaluatedl'∈S2Decision-Making Evaluation valueWherein wa(h)For the weight of h-th of evaluation index,
xlhFor value of l-th of object in h-th of evaluation index.
In order to verify the feasibility of the evaluation method of the present invention, specific experimental analysis is carried out below：
Analyzed so that South China is to third party's food inspection mechanism of government buying service as an example.According to this area
Feature, it is determined that user evaluation 7 indexs be：(1) brand requirements；(2) management requires；(3) financial requirements；(4) technology will
Ask；(5) Duty demands；(6) resource guarantee requirement；(7) legal requirement.
In order to determine the decision attribute in rough set evaluation, 10 Senior Experts are acquired first within 3 season to 30
Evaluation history data set of the individual third party's food inspection mechanism in terms of 7 indexsBy data set after pretreatment, setting
C=0.26, threshold value λ=0.85, generate Kruskal maximal trees as shown in Figure 2；Afterwards according to based on Kruskal maximal trees
Clustering realizes that step carries out cluster analysis, obtains the clustering result { 18 } of 30 third party's food inspection mechanisms,
{ 20 }, { 8,23 }, 1,2,3,4,5,6,7,9,10,11,12,13,14,15,16,17,19,21,22,24,25,26,27,28,
29,30 }.By 30 different data objects be based on this four class cluster division result be identified in respectively 0,1,2,3 numerical value it is (actual
Represent 30 third party's food inspection mechanisms in Senior Expert Decision-Making Evaluation grade it is excellent, good, in, it is poor).The cluster is drawn
Point result feeds back to Senior Expert, it was demonstrated that and this division is rational, and using this as the decision attribute (value) in follow-up decision table.
Secondly, 30 all users of third party's food inspection mechanism are gathered to be evaluated this 30 within some season pair
As user's evaluation history data set S under 7 evaluation indexes1, sample amounts to 402.Kruskal maximal trees will be based on to cluster
Divide obtained property value and be matched with data set S one by one1In each data object, and be used as its decision attribute values, construction
Decision table S is evaluated for the complete user beneficial to Rough Set Analysis.By the calculating of rough set evaluation analysis flow, 7 are obtained
The conditional information entropy importance of evaluation index is respectively 0.0001624,0.003803,0.00006960,0.001593,
0.001609,0.00005689,0.0003241.Namely obtain objective weight degree of each evaluation index under rough sets for information entropy
Value W=(0.02,0.50,0.01,0.21,0.21,0.01,0.04).It is readily seen that from the objective value result：User thinks
The management of third party's food inspection mechanism, technology, responsibility and legal requirement are even more important, food inspection mechanism should pay close attention to this four
Individual aspect.What the calculating of the objective weight was issued with State Council《Establishment of social credit system planning outline (2014-2020)》
Substantially it coincide, also proves the correctness that the objective weight based on rough set obtains to a certain extent.
Finally, in order to illustrate beneficial to user's evaluation analysis of reality, based on the evaluation method, Shenzhen is acquired in certain season
Spend to the average data of 7 third party's food inspection mechanism evaluations as test data set S2(see the table below 1), according to striked
Objective weight metric W, read group total is weighted according to the step5 of rough set evaluation analysis flow, obtains 7 inspections to be evaluated
Survey mechanism comprehensive evaluation value be respectively：91.970,89.277,81.492,85.438,85.322,86.048,84.118,
That is x1＞ x2＞ x6＞ x4＞ x5＞ x7＞ x3。
The user's evaluation test data set of table 1
In order to further prove the validity of this method, we by the present invention method (being designated as method 1) with《Examine detection
The discussion of mechanism Credit Evaluation System model》(quality technical supervision is studied, 2016,3 (45)：What is 58-60) proposed uses step analysis
Method AHP (being designated as method 2) is analyzed.
Analysis of Policy Making is carried out to the test data set in the user's evaluation example and table 1 in this method using method 2, calculated
Dominant eigenvalue λ max=7.119 are obtained, less than Critical Eigenvalue λ ' max=7.79, wherein coincident indicator CI is 0.0199,
Consistency Ratio CR is 0.015.Be derived from the AHP weights W " of 7 evaluation indexes=(0.072,0.120,0.192,0.181,
0.079,0.049,0.307).
Further, in order to verify the present invention method 1 superiority, definition scheme discrimination
WhereinN is the quantity of scheme (data object to be evaluated).Thus evaluated to calculate both the above respectively
The discrimination between scheme under method, table specific as follows：
Discrimination means that more greatly the good and bad resolvability between testing agency to be evaluated is bigger, also directly shows the present invention's
The superiority of evaluation method.
Technical scheme is not limited to the limitation of above-mentioned specific embodiment, and every technique according to the invention scheme is done
The technology deformation gone out, each falls within protection scope of the present invention.
Claims (5)
1. a kind of third party's food inspection mechanism evaluation method for government buying service, it is characterised in that including following step
Suddenly：
Step 1: gather first to multiple expert opinion history data sets for being evaluated third party's food inspection mechanismUser comments
Valency history data set S1, test data set S to be evaluated2；
Step 2: pretreatment expert opinion history data set
Step 3: calculate expert opinion history data set after pretreatmentSimilarity relation matrix, generation Kruskal is maximum
Tree, obtains the clustering result of set of data objects；
Step 4: according to data object clustering result and user's evaluation data set S of Corresponding matching1Construction evaluation decision table
S；
Step 5: ask for each user's evaluation index a ∈ C importance SGF (a, A, D) and each evaluation index in rough sets for information entropy
Under weight metric value wa；
Step 6: it is based on test data set S to be evaluated2Summation evaluation analysis is weighted with each evaluation criterion weight, is obtained to be evaluated
Valency object xl'∈S2Decision-Making Evaluation valueWherein wa(h)For the weight of h-th of evaluation index, xlhFor
Value of the l object in h-th of evaluation index.
2. third party's food inspection mechanism evaluation method according to claim 1 for government buying service, its feature
It is：To expert opinion history data set in the step 2Data prediction include：For profit evaluation model attribute, property value
According toProcessing；For cost type attribute, then according toPlace
Reason.
3. third party's food inspection mechanism evaluation method according to claim 1 for government buying service, its feature
It is：Comprised the following steps that in the step 3：
1) formula is utilizedCalculate expert opinion history data set after pretreatmentPhase
Like relational matrix R, wherein, R=(rij)n×n, c suitably chooses so that 0≤rij≤1；
2) the maximum r of value is found in similarity relation matrix R non-leading diagonalijAnd corresponding data object ri、rj, and draw
Go outLocal Kruskal maximal trees；
3) repeat step 2), it is connected to all summits, and without circle, so as to obtain overall Kruskal maximal trees；
4) it is based on to pretreated expert opinion history data setIn property value setting threshold value λ ∈ [0,1], break apart by chopping most
Connection weight r in big treeijBranch less than λ, a disconnected graph is obtained, and each connected component just constitutes the cluster in λ-level
Division result；
5) data object under same clustering is given to the mark of same integer numerical value, completes expert opinion history number
According to collectionDivision.
4. third party's food inspection mechanism evaluation method according to claim 1 for government buying service, its feature
It is：In the step 4, taken to obtaining each clustering numerical value for being evaluated third party's food inspection mechanism in step 3
Value, as the decision attribute D in decision table S and corresponding value, then gather multiple users i.e. government procurement service side by
Index i.e. conditional attribute C is determined to expert opinion history data setIncluded in third party's food inspection mechanism carry out beat
Point, i.e. user's evaluation history data set S1, by user's evaluation history data set S1In every a data object composition expert opinion
History data setWhat is obtained during cluster belong to decision attribute values of each evaluation object, structure evaluation decision table S=(U, R,
V, f) it is a decision table, wherein, U is entire objects set；R=C ∪ D are conditional attribute collection C and decision attribute D union；V
It is property set R codomain, i.e.,f:U × R → V is referred to as mapping function, i.e.,A ∈ R, there is f (xi, a)=Va。
5. third party's food inspection mechanism evaluation method according to claim 1 for government buying service, its feature
It is：The step 5 comprises the following steps：
1) set X=U/IND (Q)={ X is made1,X2,...,XpAnd Y=U/IND (P)={ Y1,Y2,...,YqIt is respectively object
Collect U derived equivalent partitions in the case where knowledge is property set P and Q, the probability distribution for defining its dividing subset is p (Xp)=| Xp|/|U
|,p(Xq)=| Yq|/| U |, wherein | | for the gesture of set；
2) conditional information entropy under knowledge Q divides relative to knowledge P is defined
Wherein, p (Yt/Xs)=| Yt∩Xs|/|Xs|；
3) basis based on more than, obtains the calculation formula of conditional attribute importance：WhereinThe then condition
Attribute a importance measures are SGF (a, A, D)=H (D/A)-H (D/A ∪ { a }), and then obtain conditional attribute a in rough set
Weight metric value under comentropy is：
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710835219.7A CN107730080A (en) | 2017-09-15 | 2017-09-15 | A kind of third party's food inspection mechanism evaluation method for government buying service |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710835219.7A CN107730080A (en) | 2017-09-15 | 2017-09-15 | A kind of third party's food inspection mechanism evaluation method for government buying service |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN107730080A true CN107730080A (en) | 2018-02-23 |
Family
ID=61206560
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710835219.7A Pending CN107730080A (en) | 2017-09-15 | 2017-09-15 | A kind of third party's food inspection mechanism evaluation method for government buying service |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107730080A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116664152A (en) * | 2023-06-15 | 2023-08-29 | 浙江智飨科技有限公司 | Supply chain full-flow node tracing system and method |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2015092193A1 (en) * | 2013-12-20 | 2015-06-25 | Electricite De France | Optimised energy supply to a distribution network |
| CN105893748A (en) * | 2016-03-29 | 2016-08-24 | 安徽大学 | Happiness index evaluation method and system based on rough set theory |
| CN106845142A (en) * | 2017-03-09 | 2017-06-13 | 南京大学 | Quality evaluation method based on improved rough set Set Pair Analysis |
-
2017
- 2017-09-15 CN CN201710835219.7A patent/CN107730080A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2015092193A1 (en) * | 2013-12-20 | 2015-06-25 | Electricite De France | Optimised energy supply to a distribution network |
| CN105893748A (en) * | 2016-03-29 | 2016-08-24 | 安徽大学 | Happiness index evaluation method and system based on rough set theory |
| CN106845142A (en) * | 2017-03-09 | 2017-06-13 | 南京大学 | Quality evaluation method based on improved rough set Set Pair Analysis |
Non-Patent Citations (1)
| Title |
|---|
| 高维春: "决策属性未知下的学生评教粗糙集分析", 《计算机工程与应用》 * |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116664152A (en) * | 2023-06-15 | 2023-08-29 | 浙江智飨科技有限公司 | Supply chain full-flow node tracing system and method |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Gang et al. | The contribution of marketization to China's economic growth | |
| Li et al. | Sustainable supplier selection for water environment treatment public-private partnership projects | |
| CN108846592A (en) | A kind of valuation of enterprise report-generating method and device based on big data | |
| Naeini et al. | A hybrid model of competitive advantage based on Bourdieu capital theory and competitive intelligence using fuzzy Delphi and ism-gray Dematel (study of Iranian food industry) | |
| Hossain et al. | The relationship between corporate social responsibility and corporate financial performance: Evidence from a developing country | |
| Wang et al. | The empirical study on operating efficiency of agricultural cooperatives in Langao | |
| Distefano et al. | Sub-national water–food–labour nexus in Colombia | |
| Rastpour et al. | Multi-criteria decision-making methods for the evaluating of a real green supply chain in companies with fast-moving consumer goods | |
| Balcı et al. | Hierarchies in communities of UK stock market from the perspective of Brexit | |
| Li et al. | Digital inclusive finance & the high-quality agricultural development: Prevalence of regional heterogeneity in rural China | |
| Cong et al. | New smart city clusters’ construction level evaluation under economic circles: The case of Shandong, China | |
| Ma et al. | Does land transfer have an impact on land use efficiency? A case study on rural China | |
| CN107730080A (en) | A kind of third party's food inspection mechanism evaluation method for government buying service | |
| KR101259417B1 (en) | Hybrid type method and system for extracting a emerging technologies using collective intelligence | |
| Xu et al. | Ranking academic impact of world national research institutes—by the Chinese Academy of Sciences | |
| An et al. | Performance measurement of decision making units through interval efficiency with slacks-based measure: an application to tourist hotels in Taipei | |
| Wang et al. | An analysis of the Economic Impact of US Presidential elections based on principal component and logical regression | |
| Wu et al. | Chinese airline competitiveness evaluation based on extended binary relative evaluation (BRE) model | |
| CN114511174A (en) | Service index map construction method and device | |
| Mu et al. | Effects of Environmental Regulation Competition and Public Participation on Enterprise Location Selection Under Climate Change in COVID-19 Pandemic Conditions: An Analysis Based on the Chinese Provincial Spatial Panel Model | |
| Zhu et al. | Evaluation and Analysis of Land Input‐Output Comprehensive Benefit Based on Fuzzy Mathematics and Analytic Hierarchy Process | |
| Yan et al. | Portfolio optimization: A return-on-equity network analysis | |
| Erjavec et al. | SMEs “Growing Smart”: The Complementarity of Intangible and Digital Investment in Small Firms and Their Contribution to Firm Performance | |
| Wang | Fuzzy evaluations applied in analysis of infrastructure development based on national competitiveness perspectives | |
| Sun et al. | How does digital finance promote entrepreneurship? The roles of traditional financial institutions and BigTech firms |
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 | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20180223 |