CN114897281A - Expert portrait calculation method - Google Patents
Expert portrait calculation method Download PDFInfo
- Publication number
- CN114897281A CN114897281A CN202210000424.2A CN202210000424A CN114897281A CN 114897281 A CN114897281 A CN 114897281A CN 202210000424 A CN202210000424 A CN 202210000424A CN 114897281 A CN114897281 A CN 114897281A
- Authority
- CN
- China
- Prior art keywords
- expert
- influence
- experts
- score
- direct influence
- 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
- 238000004364 calculation method Methods 0.000 title claims abstract description 7
- 239000011159 matrix material Substances 0.000 claims abstract description 48
- 238000000034 method Methods 0.000 claims abstract description 35
- 238000011156 evaluation Methods 0.000 claims abstract description 12
- 230000000694 effects Effects 0.000 claims description 6
- 238000012512 characterization method Methods 0.000 claims description 4
- 230000007246 mechanism Effects 0.000 claims description 3
- 230000008520 organization Effects 0.000 claims description 2
- 238000005259 measurement Methods 0.000 abstract description 7
- 230000008569 process Effects 0.000 abstract description 5
- 238000012360 testing method Methods 0.000 abstract description 3
- 238000010998 test method Methods 0.000 abstract 1
- 238000013461 design Methods 0.000 description 2
- 238000011005 laboratory method Methods 0.000 description 2
- 238000011160 research Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000010606 normalization Methods 0.000 description 1
- 238000012827 research and development Methods 0.000 description 1
- 238000012795 verification Methods 0.000 description 1
Images
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
Landscapes
- Business, Economics & Management (AREA)
- Human Resources & Organizations (AREA)
- Engineering & Computer Science (AREA)
- Strategic Management (AREA)
- Development Economics (AREA)
- Economics (AREA)
- Entrepreneurship & Innovation (AREA)
- Educational Administration (AREA)
- Operations Research (AREA)
- Marketing (AREA)
- Game Theory and Decision Science (AREA)
- Quality & Reliability (AREA)
- Tourism & Hospitality (AREA)
- Physics & Mathematics (AREA)
- General Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses an expert image computing method, which is a computing method with the core of direct influence matrix of expert relationship. Firstly, establishing an evaluation index system of direct influence relationship among experts; then, the influence score of each index on the direct influence relation is calculated in a pairwise hierarchical comparison mode, the comprehensive measurement of the direct influence relation among experts is obtained in a weighted summation mode, and a direct influence matrix is constructed according to the comprehensive measurement; and then solving the centrality and the reason degree of each expert to verify the reasonability of the direct influence matrix. The invention comprehensively considers the common attribute between every two experts and the specific attribute of the experts in the process of constructing the direct influence matrix of the experts, solves the problems that the decision test and evaluation test method (DEMATEL) method directly constructs the direct influence matrix in application, does not consider the complexity of a decision environment and the uncertainty of the judgment information of the experts, and effectively improves the applicability of the DEMATEL method.
Description
Technical Field
The invention relates to the field of expert portrait research, and particularly provides a method for calculating an expert relation direct influence matrix.
Background
In the expert portrait research, the Decision of a Decision test and Evaluation Laboratory method (Decision Making and Evaluation Laboratory) on a problem is judged based on the direction and degree of the pairwise influence relationship of the experts, and the correlation relationship between the experts and the measurement of the influence degree can be calculated by constructing a direct influence matrix, so that the structural relationship of the problem is known. The structural relation plays a crucial role in the research and development process of design strategies.
The invention comprehensively considers the common attribute between every two experts and the specific attribute of the experts, solves the problems that the DEMATEL method directly constructs a direct influence matrix in application and does not consider the complexity of a decision environment and the uncertainty of the judgment information of the experts, helps researchers in the design field to better understand and apply a decision test and evaluation laboratory method, and effectively improves the applicability of the DEMATEL method.
Disclosure of Invention
The invention provides a calculation method for directly influencing a matrix by an expert relationship in order to solve the problems mentioned in the background, fully consider the common attribute between every two experts and the specific attribute of the experts.
Firstly, establishing an evaluation index system of direct influence relationship among experts; then, the influence score of each index on the direct influence relation is calculated in a pairwise layered comparison mode, the comprehensive measurement of the direct influence relation among experts is obtained in a weighted summation mode, and a direct influence matrix is constructed according to the comprehensive measurement; and then solving the centrality and the reason degree of each expert to verify the reasonability of the direct influence matrix.
1. A method for calculating an expert relationship direct influence matrix is characterized by comprising the following steps: the method comprises the following steps:
(1) and creating an evaluation index system of which the relation is directly influenced by the expert. The indexes can be divided into two types, one type is a common attribute between every two experts, and the index comprises the following steps: number of commonly published articles x 1 From the same mechanism x 2 (including primary and secondary units), participating in a common item quantity x 3 Etc.; another class is the characteristic attributes of the expert itself, including: total number of publications x 4 Reference number (position) x 5 Age x 6 Job title x 7 Etc.;
(2) converting each index variable into a quantitative value by adopting a pairwise hierarchical comparison method, then respectively calculating a total score of the influence of the common attributes and a total score of the influence of the specific attributes, and determining the connection relation and the arrow direction between experts according to the total scores;
(3) quantifying the comprehensive direct influence among experts by calculating the weighted sum of all indexes and constructing a direct influence matrix according to the comprehensive direct influence;
(4) and solving the centrality and the reason degree of each expert, and verifying the rationality of the direct influence matrix.
2. The method of claim 1, wherein the method comprises: in the step (2), x is { x ═ x i The evaluation of the value of i ═ 1,2,3} can be obtained as follows:
wherein: s 1 Representing the effect of the number of commonly published articles on the expert relations. m is 1 The average value of the number of the commonly published articles in the expert data set is calculated when the number of the commonly published articles reaches m 1 In the first time, two experts are considered to have stronger relationship scores. s is 1 The number of the commonly published articles is divided into 3 grades, and the values are 0 and s respectively 1_1 ＝3，s 1_2 ＝5。
s 2 Representing the effect on expert relations whether they come from the same organization. Final score s 2 Dividing into 3 grades from different units, the same primary unit and the same secondary unit, and respectively taking the score of 0 and s 2_1 ＝3，s 2_2 ＝5。
s 3 Representing the effect of the number of co-participating items on the expert relationship. m is 3 Is the average value of the number of the commonly participated projects in the expert data set when the number of the commonly participated projects reaches m 2 At one time, two experts are considered to have a strong relationship with a score of s 3_1 (ii) a Final score s 3 The number of the commonly published articles is divided into 3 grades, and the values are 0 and s respectively 3_1 ＝3，s 3_2 ＝5。
The influence weight of the common attribute on the correlation relationship between the experts is recorded as lambda i (i is 1,2,3), the relationship between expert i and expert j
Can be characterized as:
finally according to
Judging the connection relationship between every two experts if
If the value is 0, no connection exists, otherwise, connection exists.
3. The method of claim 1, wherein the method comprises: the characteristic attributes of the expert in the step (2) comprise: total number of publications x 4 Reference number (position) x 5 Age x 6 Job title x 7 And the like.
Wherein the total number of publications x 4 Reference number (position) x 5 The score is calculated by:
wherein s is 4 A score representing the total number of publications. m is 4_1 Value of 1/4 quantile, m, sorted according to expert total article posting dataset 4_2 The value of 3/4 quantile points is sorted by the total number of the expert issued chapters; s 5 Representing the score of the citation. m is 5_1 1/4 quantile values, m, sorted according to different article quotation data sets 5_2 The article quotes the value of 3/4 quantile in order.
Age x 6 Heyu title x 7 A scale reference needs to be constructed from which an influence score is given. Wherein the ages are divided into 5 grades with the scores of 1-5. The titles are divided into 3 grades, and the scores are 1, 3 and 5 respectively.
Age x 6 Heyu title x 7 Need to constructThe table is referenced to a scale from which an influence score is given.
| Age (age) | Scale of influence | Score of |
| Less than 30 | Is very small | 1 |
| 30-40 | Small | 2 |
| 40-50 | Medium and |
3 |
| 50-60 | Big (a) | 4 |
| More than 60 | Is very large | 5 |
| Job scale | Scale of influence | Score of |
| Lecturer/instructor | Is very small | 1 |
| Assistant professor/assistant investigator | |
3 |
| Institute/professor | Is very large | 5 |
The weight of the degree of influence of the characteristic attribute on the expert is recorded as lambda j (j＝4,5,6,7)。
The degree of influence s of expert i on expert j i→j And the degree of influence s of expert i on expert j j→i Can be characterized as:
in the formula, superscripts i and j denote the expert reference symbols that exert the influence.
4. The method of claim 1, wherein the method comprises: the integrated direct influence among the experts in said step (3) may be characterized as the sum of the interrelationship among the experts and the degree of influence of the expert on another expert.
Then expert i is right to expertComprehensive direct influence s of Home j ij The final characterization was:
the combined direct influence s of expert i on expert j ji The final characterization was:
finally according to s ij And s ji The filling directly affects the matrix S as follows:
5. in step (4), the standard influence matrix N (nxn) is calculated according to the direct influence matrix
Then
Derived from Newman's formula
T＝N(I-N) -1
Wherein T is a comprehensive influence matrix, N is a normalized influence matrix, I is an identity matrix,
(I-N) -1 is the inverse matrix of I-N.
The influence degree refers to the sum of the influence of any one expert on other experts, and the set is marked as D.
D＝(D 1 ,D 2 ,D 3 ,…,D n )
In the formula, t ij The degree of the direct influence and the indirect influence brought by the expert i to the expert j is represented, namely the comprehensive influence degree is generated. And also indicates the comprehensive influence degree of the expert j by the expert i.
The degree of influence refers to the sum of the influence exerted by other experts on any one expert, and the set is denoted as C.
C＝(C 1 ,C 2 ,C 3 ,…,C n )
Wherein,
the centrality refers to the influence degree of the expert i and the influence degree are added to obtain the centrality of the expert, which is recorded as M i . Centrality represents the position of the expert in the expert decision system and the size of the contribution it plays.
M i ＝D i +C i
The reason degree refers to the degree of influence of the expert i and the degree of influence are subtracted to obtain the centrality of the expert, which is recorded as R i . Centrality represents the position of the expert in the expert decision system and the size of its contribution.
R i ＝D i -C i
The relevance of the decision of each expert and the status of each expert in the decision process can be judged and determined through the centrality and the reason.
The technical scheme of the invention has the following beneficial effects:
in the scheme, the method comprises the steps of firstly constructing an evaluation index system of direct influence relation among experts; then, the influence score of each index on the direct influence relation is calculated in a pairwise hierarchical comparison mode, the comprehensive measurement of the direct influence relation among experts is obtained in a weighted summation mode, and a direct influence matrix is constructed according to the comprehensive measurement; and then solving the centrality and the reason degree of each expert to verify the reasonability of the direct influence matrix. The invention comprehensively considers the common attribute between every two experts and the specific attribute of the experts in the process of constructing the direct influence matrix of the experts, solves the problems that the direct influence matrix is directly constructed in the application of the DEMATEL method, the complexity of a decision environment and the uncertainty of the judgment information of the experts are not considered, and effectively improves the applicability of the DEMATEL method.
Drawings
FIG. 1 is a flow chart of a method of computing an expert relationship direct impact matrix of the present invention;
FIG. 2 is a directed graph
Detailed Description
In order to make the technical problems, technical solutions and advantages of the present invention more apparent, the following detailed description is given with reference to the accompanying drawings and specific embodiments.
1. A method for calculating an expert relationship direct influence matrix is characterized by comprising the following steps: the method comprises the following steps:
(1) and creating an evaluation index system of which the relation is directly influenced by the expert. The indexes can be divided into two types, one type is the common attribute between every two experts, and the method comprises the following steps: number of commonly published articles x 1 From the same mechanism x 2 (including primary and secondary units), participating in a common item quantity x 3 Etc.; another class is the characteristic attributes of the expert itself, including: total number of publications x 4 Reference number (position) x 5 Age x 6 Job title x 7 Etc.;
(2) converting each index variable into a quantitative value by adopting a pairwise hierarchical comparison method, then respectively calculating a total score of the influence of the common attributes and a total score of the influence of the specific attributes, and determining the connection relation and the arrow direction between experts according to the total scores;
expert data in a certain field is crawled from an AMIner website and an official website of the science foundation committee of natural sciences, which are created by a Qinghua university team, and statistics is carried out on characteristics of a data set. And 5 experts are selected for verification.
Wherein, the score statistical table of the common attributes is shown in the following tables 1-4.
TABLE 1 score statistics table for the number of articles published by experts together
| Expert 1 | Expert 2 | |
Expert 4 | |
|
| Expert 1 | - | 5 | 5 | 3 | 3 |
| Expert 2 | 5 | - | 3 | 3 | 0 |
| |
5 | 3 | - | 3 | 5 |
| Expert 4 | 3 | 3 | 3 | - | 3 |
| |
3 | 0 | 5 | 3 | - |
The values in the table above are each a measure of the influence score of expert i on expert j,
TABLE 2 score statistics of whether experts are from the same unit
| Expert 1 | Expert 2 | |
Expert 4 | |
|
| Expert 1 | - | 5 | 0 | 5 | 0 |
| Expert 2 | 5 | - | 0 | 3 | 0 |
| |
0 | 0 | - | 0 | 5 |
| Expert 4 | 5 | 3 | 0 | - | 0 |
| |
0 | 0 | 5 | 0 | - |
TABLE 3 statistics table for scores of items commonly implemented by experts
| Expert 1 | Expert 2 | |
Expert 4 | |
|
| Expert 1 | - | 5 | 0 | 3 | 0 |
| Expert 2 | 5 | - | 0 | 3 | 0 |
| |
0 | 0 | - | 3 | 5 |
| Expert 4 | 3 | 3 | 3 | - | 0 |
| |
0 | 0 | 5 | 0 | - |
The total score of the common attributes between expert i and expert j can be characterized as:
from practical experience, λ 1 ＝λ 2 ＝λ 3 ＝1。
TABLE 4 Total score of common attributes among experts
Matrix statistical table
| Expert 1 | Expert 2 | |
Expert 4 | |
|
| Expert 1 | - | 15 | 5 | 11 | 3 |
| Expert 2 | 15 | - | 3 | 9 | 5 |
| |
5 | 3 | - | 6 | 15 |
| Expert 4 | 11 | 9 | 6 | - | 3 |
| |
3 | 5 | 15 | 3 | - |
The characteristic attribute score statistics are shown in tables 5 and 6 below,
TABLE 5 expert-specific Attribute valuating Table
| X 4 | X 5 | X 6 | X 7 | |
| Expert 1 | 399 | 4629 | 58 | Teaching of |
| Expert 2 | 195 | 4092 | 43 | |
| Expert | ||||
| 3 | 444 | 4318 | 63 | Teaching of |
| Expert 4 | 84 | 1662 | 33 | |
| Expert | ||||
| 5 | 109 | 2108 | 40 | Secondary investigator |
TABLE 6 expert peculiar attribute value scoring table
| S 4 | S 5 | S 6 | S 7 | Total score | |
| Expert 1 | 5 | 5 | 4 | 5 | 19 |
| Expert 2 | 3 | 4 | 3 | 3 | 13 |
| |
5 | 5 | 5 | 5 | 20 |
| Expert 4 | 2 | 2 | 1 | 1 | 6 |
| |
3 | 3 | 2 | 5 | 13 |
A directed graph of the direct influence between experts can be drawn according to tables 4 and 6, as shown in FIG. 2, wherein A to E represent experts 1 to 5, respectively, the numbers in the nodes are attribute scores unique to each expert, and the numbers on the sides are attribute scores common to the experts.
(3) Quantifying the comprehensive direct influence among experts by calculating the weighted sum of all indexes and constructing a direct influence matrix according to the comprehensive direct influence;
the overall direct influence of expert i on expert j is finally characterized as:
the overall direct influence of expert i on expert j is finally characterized as:
the above formula is expressed in matrix form, see table 7.
TABLE 7 Total scores of common and unique attributes among experts
Matrix statistical table
| Expert 1 | Expert 2 | |
Expert 4 | |
|
| Expert 1 | - | 34 | 24 | 30 | 22 |
| Expert 2 | 28 | - | 16 | 21 | 18 |
| |
25 | 23 | - | 26 | 35 |
| Expert 4 | 17 | 15 | 12 | - | 9 |
| |
16 | 18 | 28 | 16 | - |
(4) And solving the centrality and the reason degree of each expert by using a DEMETAL method, and verifying the rationality of the direct influence matrix.
Quantifying the correlation between expert i and expert j by calculating the weighted sum of the indexes and constructing S according to the correlation N×N Directly affects the matrix.
Calculating a standard influence matrix and then calculating normalization to obtain:
TABLE 8 normalized Specification impact matrix statistics Table
| Expert 1 | Expert 2 | |
Expert 4 | |
|
| Expert 1 | 0 | 0.309 | 0.218 | 0.273 | 0.200 |
| Expert 2 | 0.255 | 0 | 0.145 | 0.191 | 0.164 |
| |
0.227 | 0.209 | 0 | 0.236 | 0.318 |
| Expert 4 | 0.155 | 0.136 | 0.109 | 0 | 0.082 |
| |
0.145 | 0.164 | 0.255 | 0.145 | 0 |
From the newman formula:
T＝N(I-N) -1
wherein T is a comprehensive influence matrix, N is a normalized influence matrix, I is an identity matrix,
(I-N) -1 is the inverse matrix of I-N.
TABLE 9 statistical table of comprehensive impact matrix
| Expert 1 | Expert 2 | |
Expert 4 | |
Influence degree D | |
| Expert 1 | 0.714 | 0.977 | 0.839 | 0.976 | 0.849 | 4.354 |
| Expert 2 | 0.777 | 0.597 | 0.662 | 0.774 | 0.691 | 3.501 |
| |
0.892 | 0.906 | 0.668 | 0.946 | 0.935 | 4.347 |
| Expert 4 | 0.524 | 0.526 | 0.461 | 0.418 | 0.453 | 2.382 |
| |
0.680 | 0.711 | 0.722 | 0.716 | 0.540 | 3.368 |
| Degree of influence C | 3.586 | 3.717 | 3.351 | 3.830 | 3.469 | - |
The influence degree refers to the sum of the influence of any one expert on other experts, and the set is marked as D.
D＝(D 1 ,D 2 ,D 3 ,…,D n )
In the formula, t ij The degree of the direct influence and the indirect influence brought by the expert i to the expert j is represented, namely the comprehensive influence degree is generated. And also indicates the comprehensive influence degree of the expert j by the expert i.
The degree of influence refers to the sum of the influence exerted by other experts on any one expert, and the set is denoted as C.
C＝(C 1 ,C 2 ,C 3 ,…,C n )
Wherein,
the centrality refers to the influence degree of the expert i and the influence degree are added to obtain the centrality of the expert, which is recorded as M i . Centrality represents the position of the expert in the expert decision system and the size of its contribution.
M i ＝D i +C i
The reason degree refers to the degree of influence of the expert i and the degree of influence are subtracted to obtain the centrality of the expert, which is recorded as R i . Centrality represents the position of the expert in the expert decision system and the size of the contribution it plays.
R i ＝D i -C i
The relevance of the decision of each expert and the status of each expert in the decision process can be judged and determined through the centrality and the reason.
TABLE 10 centrality and reason result statistics table
| Influence degree D | Degree of influence C | Center degree M | Degree of cause R | |
| Expert 1 | 4.354 | 3.586 | 7.940 | 0.768 |
| Expert 2 | 3.501 | 3.717 | 7.218 | -0.216 |
| |
4.347 | 3.351 | 7.698 | 0.996 |
| Expert 4 | 2.382 | 3.830 | 6.212 | -1.448 |
| |
3.368 | 3.469 | 6.837 | -0.100 |
From the above table, it can be seen that the centrality of expert 1 and expert 3 is greater, which indicates that expert 1 and expert 3 have a more important status in the decision system, and have a wider connection and are more active in the expert decision system. The cause degree of experts 1, 3 is a large positive value, which indicates that the cause factor is in the decision system, and the cause degree of experts 2, 4,5 is a negative value, and the cause factor is in the decision system.
While the foregoing is directed to the preferred embodiment of the present invention, it will be understood by those skilled in the art that various changes and modifications may be made without departing from the spirit and scope of the invention as defined in the appended claims.
Claims (4)
1. An expert portrait calculation method is mainly a calculation method of an expert relationship direct influence matrix, and is characterized in that:
(1) and creating an evaluation index system of which the relation is directly influenced by the expert. The indexes can be divided into two types, one type is a common attribute between every two experts, and the index comprises the following steps: number of commonly published articles x 1 From the same mechanism x 2 (including primary and secondary units), participating in a common item quantity x 3 Etc.; another class is the characteristic attributes of the expert itself, including: total number of publications x 4 Reference number (position) x 5 Age x 6 Job title x 7 Etc.;
(2) converting each index variable into a quantitative value by adopting a pairwise hierarchical comparison method, then respectively calculating a total score of the influence of the common attributes and a total score of the influence of the specific attributes, and determining the connection relation and the arrow direction between experts according to the total scores;
(3) quantifying the comprehensive direct influence among experts by calculating the weighted sum of all indexes and constructing a direct influence matrix according to the comprehensive direct influence;
(4) and solving the centrality and the reason degree of each expert, and verifying the rationality of the direct influence matrix.
2. The method of claim 1, wherein the method comprises: in the step (2), x is { x ═ x i The evaluation of the value of i ═ 1,2,3} can be obtained as follows:
wherein: s 1 Representing the effect of the number of commonly published articles on the expert relations. m is 1 For collective sending of expert dataThe average value of the number of the published articles is when the number of the commonly published articles reaches m 1 In the first place, two experts are considered to have stronger relationship scores. s 1 The number of the commonly published articles is divided into 3 grades, and the values are 0 and s respectively 1_1 ＝3，s 1_2 5. The score matrix of the number of commonly published articles is expressed as (S) 1 ) n×n ，(S 1 ) ij Is s is 1 Represents the score of the common attribute of the expert i and the expert j.
s 2 Representing the effect on expert relations whether they come from the same organization. Final score s 2 Dividing into 3 grades from different units, the same primary unit and the same secondary unit, and respectively taking the score of 0 and s 2_1 ＝3，s 2_2 5. The score matrix from the same unit is denoted S 2 ，(S 2 ) ij Is s is 2 Represents the score of the common attribute of the expert i and the expert j.
s 3 Representing the effect of the number of co-participating items on the expert relationship. m is 3 Is the average value of the number of the common participation items in the expert data set when the number of the common participation items reaches m 3 At one time, two experts are considered to have a strong relationship with a score of s 3_1 (ii) a Final score s 3 The number of the commonly published articles is divided into 3 grades, and the values are 0 and s respectively 3_1 ＝3，s 3_2 5. The scoring matrix of the co-participating items is denoted as S 3 ，(S 3 ) ij Is s is 3 Represents the score of the common attribute of the expert i and the expert j.
The influence weight of the common attribute on the correlation relationship between the experts is recorded as lambda i (i is 1,2,3), the common attribute integrated influence matrix can be recorded as
Wherein the mutual influence relationship between expert i and expert j
Can be characterized as:
finally according to
Judging the connection relationship between every two experts if
If 0, there is no connection, otherwise there is connection.
3. The method of claim 1, wherein the method comprises: the specific attributes of the experts in the step (2) comprise: total number of publications x 4 Reference number (position) x 5 Age x 6 Job title x 7 And the like.
Wherein the total number of publications x 4 Reference number (position) x 5 The score is calculated by:
wherein s is 4 A score representing the total number of publications. m is 4_1 Value of 1/4 quantile, m, sorted according to expert total article posting dataset 4_2 The value of 3/4 quantile points is sorted by the total number of the expert issued chapters; s 5 Representing the score of the citation. m is 5_1 1/4 quantile values, m, sorted according to different article quotation data sets 5_2 The article quotes the value of 3/4 quantile in order.
Age x 6 Heyu title x 7 A scale reference needs to be constructed from which an influence score is given. Wherein the ages are divided into 5 grades with the scores of 1-5. The titles are divided into 3 grades, and the scores are 1, 3 and 5 respectively.
Characteristic attribute to expertThe weight of the degree of influence is denoted as λ j (j-4, 5,6, 7). The degree of influence s of expert i on expert j i→j And degree of influence s of expert i and expert j j→i Can be characterized as follows:
in the formula, superscripts i and j denote the expert reference symbols that exert the influence.
4. The method of claim 1, wherein the method comprises: the integrated direct influence among the experts in said step (3) may be characterized as the sum of the interrelationship among the experts and the degree of influence of the expert on another expert.
The combined direct influence s of expert i on expert j ij The final characterization was:
the combined direct influence s of expert i on expert j ji The final characterization was:
finally according to s ij And s ji The filling directly affects the matrix S as follows:
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210000424.2A CN114897281A (en) | 2022-01-04 | 2022-01-04 | Expert portrait calculation method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210000424.2A CN114897281A (en) | 2022-01-04 | 2022-01-04 | Expert portrait calculation method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114897281A true CN114897281A (en) | 2022-08-12 |
Family
ID=82715412
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210000424.2A Pending CN114897281A (en) | 2022-01-04 | 2022-01-04 | Expert portrait calculation method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114897281A (en) |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102880657A (en) * | 2012-08-31 | 2013-01-16 | 电子科技大学 | Expert recommending method based on searcher |
| CN107194672A (en) * | 2016-11-09 | 2017-09-22 | 北京理工大学 | It is a kind of to merge academic speciality and the evaluation distribution method of community network |
| CN110688405A (en) * | 2019-08-23 | 2020-01-14 | 上海科技发展有限公司 | Expert recommendation method, device, terminal and medium based on artificial intelligence |
| CN110781368A (en) * | 2019-10-22 | 2020-02-11 | 北京赛时科技有限公司 | Information crawling system and method for specified experts |
| CN111008215A (en) * | 2019-11-29 | 2020-04-14 | 中科院计算技术研究所大数据研究院 | Expert recommendation method combining label construction and community relation avoidance |
-
2022
- 2022-01-04 CN CN202210000424.2A patent/CN114897281A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102880657A (en) * | 2012-08-31 | 2013-01-16 | 电子科技大学 | Expert recommending method based on searcher |
| CN107194672A (en) * | 2016-11-09 | 2017-09-22 | 北京理工大学 | It is a kind of to merge academic speciality and the evaluation distribution method of community network |
| CN110688405A (en) * | 2019-08-23 | 2020-01-14 | 上海科技发展有限公司 | Expert recommendation method, device, terminal and medium based on artificial intelligence |
| CN110781368A (en) * | 2019-10-22 | 2020-02-11 | 北京赛时科技有限公司 | Information crawling system and method for specified experts |
| CN111008215A (en) * | 2019-11-29 | 2020-04-14 | 中科院计算技术研究所大数据研究院 | Expert recommendation method combining label construction and community relation avoidance |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Yin et al. | Does e-commerce narrow the urban–rural income gap? Evidence from Chinese provinces | |
| Chen et al. | Cross-efficiency aggregation method based on prospect consensus process | |
| US8303310B2 (en) | System for evaluating information competency and method thereof | |
| CN101887460A (en) | Document quality assessment method and application | |
| Tong et al. | New peer effect-based approach for service matching in cloud manufacturing under uncertain preferences | |
| CN114139940A (en) | Generalized demand side resource network load interaction level assessment method based on combined empowerment-cloud model | |
| Li et al. | A hybrid model for experts finding in community question answering | |
| CN111523768A (en) | Entropy weight-TOPSIS-based generalized demand side resource quality evaluation method | |
| Noh et al. | Power users are not always powerful: The effect of social trust clusters in recommender systems | |
| Strutynska et al. | Small and medium business structures clustering method based on their digital maturity | |
| Liang et al. | Big and strong or small and beautiful: Effects of organization size on the performance of farmer cooperatives in China | |
| CN110942830A (en) | Background management system for psychological evaluation | |
| CN114897281A (en) | Expert portrait calculation method | |
| Liu et al. | On novel peer review system for academic journals: Analysis based on social computing | |
| Alzaghal et al. | Moderating effect of information and communication technology tools on the relationship between networking services and incubator success | |
| Lee et al. | Coherence analysis of research and education using topic modeling | |
| Organisciak et al. | Evaluating rater quality and rating difficulty in online annotation activities | |
| Hongyun | Application of data mining technology based on improved neural network algorithm in sports performance prediction | |
| CN114826967B (en) | Information sharing capability evaluation method and device | |
| UGWUDE et al. | UNIVERSITY ADMINISTRATORS’USE OF MANAGEMENT INFORMATION SYSTEMS (MIS) FEEDBACK MECHANISMS FOR DECISION-MAKING IN SOUTH EAST NIGERIAN UNIVERSITIES | |
| Solow et al. | Improved heuristics for finding balanced teams | |
| CN102768751A (en) | Learning resource push system and learning resource push method | |
| Damiri et al. | Application of Profile Matching in Selection of the Best Marketing at PT. Dinara Semesta Propertindo | |
| Cai et al. | Cooperative research and development behavior in technology-based small enterprises: The formation mechanism | |
| CN113408089B (en) | Inter-cluster influence modeling method based on gravitational field idea and storage medium |
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 |