CN108510205A - A kind of author's technical capability evaluation method based on hypergraph - Google Patents
A kind of author's technical capability evaluation method based on hypergraph Download PDFInfo
- Publication number
- CN108510205A CN108510205A CN201810316651.XA CN201810316651A CN108510205A CN 108510205 A CN108510205 A CN 108510205A CN 201810316651 A CN201810316651 A CN 201810316651A CN 108510205 A CN108510205 A CN 108510205A
- Authority
- CN
- China
- Prior art keywords
- author
- technical ability
- field
- distance
- paper
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000011156 evaluation Methods 0.000 title claims abstract description 14
- 238000011160 research Methods 0.000 claims abstract description 25
- 238000000034 method Methods 0.000 claims abstract description 19
- 230000008859 change Effects 0.000 claims abstract description 15
- 230000008569 process Effects 0.000 claims abstract description 5
- 238000004364 calculation method Methods 0.000 claims description 8
- 238000010606 normalization Methods 0.000 claims description 4
- 238000012545 processing Methods 0.000 claims description 4
- 238000001228 spectrum Methods 0.000 claims description 3
- 238000005303 weighing Methods 0.000 claims description 3
- 238000012935 Averaging Methods 0.000 claims 2
- 238000005259 measurement Methods 0.000 abstract description 2
- 239000002994 raw material Substances 0.000 abstract description 2
- 230000004069 differentiation Effects 0.000 description 4
- 238000011161 development Methods 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000005266 casting Methods 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 239000004744 fabric Substances 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 238000012360 testing method Methods 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
- G06Q10/06398—Performance of employee with respect to a job function
Abstract
The invention belongs to scholar's technical capability evaluation technical fields, are related to a kind of scholar's technical capability evaluation method based on hypergraph, can fine-grained assessment scholar in the level of a certain technical ability in a certain field, and scholar's technical ability can be embodied and changed with time rule.This method considers the factors such as paper number and Quality of Papers, the otherness of different field, time change.The use of hypergraph concept allows this method to merge scholar, field and technical ability, and to make, process provides a kind of fine-grained evaluation schemes.Calculate scholar, field and technical ability apart from when, expanded based on the Traditional measurements parameters such as paper reference amount, H index, ensure that reliability, while normalized use improves operation efficiency, reduces error.Being eventually adding time factor allows this method to analyze scholar field, and technical ability changes with time, and is provided more " raw material " for research.
Description
Technical field
The invention belongs to author's technical capability evaluation technical fields, are related to a kind of author's technical capability evaluation method based on hypergraph.
Background technology
With the continuous development of science and technology, the author's number being engaged in scientific research is more and more, and the growth of scientific research personnel's quantity promotees
The research to scientific research personnel is made.The level for assessing a scientific research personnel, there is which speciality, paper publishing have which rule
Deng, the establishment to Research Team, project investment, the comparison of the academic level of global scope assessment author, different authors, research science
Cooperation behavior mechanism finds that the potential rule etc. of scientific research has facilitation, is conducive to academia or even human society
Progress and development.
H-index is mostly used greatly to the assessment of author's level at present, paper is by argument amount, the indexs such as paper publishing quantity.
These parameters are typically to carry out total evaluation to an author, and a certain skill proficiency of author can not be known, can not know by existing
The problems such as academic level variation of author's certain time period, the range and depth of research are limited to a certain extent.
Hypergraph is a kind of extensive figure, and the super side of one item can include multiple vertex.The characteristics of super side, allows hypergraph to melt
The polynary attribute of partner extremely meets the author required in author's technical ability research, field, the processing of three element of technical ability.
Invention content
The purpose of the present invention proposes a kind of author technical capability evaluation side mainly for some shortcomings of above-mentioned existing research
Method, by analyzing contribution of the author in publishing thesis, binding time factor proposes that a kind of author based on hypergraph comments
Estimation algorithm.The algorithm has carried out fine-grained assessment to author, it is contemplated that paper number and Quality of Papers, it is also considered that difference neck
The otherness in domain, by this algorithm can know author a certain specific technical ability in a certain field proficiency, while when adding
Between factor, author's technical ability can be obtained and changed with time.
Technical scheme of the present invention：
A kind of author's technical capability evaluation method based on hypergraph, steps are as follows：
Step 1)：By the field super side of composition of the technical ability and this paper of Authors of Science Articles, author in that article, author is counted
All papers participated in, merger kinds of skill obtain Authors of Science Articles, technical ability and the statistical data in field；Hypergraph has good
Compatibility can merge three author, field and technical ability factors；The paper information that author delivers is counted, author, technical ability
Three vertex with field as super side；
A certain technical ability of the super side by author in some field connects, and the weight by calculating super side embodies author at certain
The proficiency of a certain technical ability in one field, while can effectively reduce using hypergraph author's network size of research；
Since technical ability is complicated, the correlation computations of later step can be influenced, the technical ability in data set is subjected to merger, is united
One standard data set；
Step 2)：Three vertex on super side are combined two-by-two, calculate the distance on each vertex in hypergraph；Between each attribute
Distance is calculated by following formula：
Author j is at a distance from the f of field：
Wherein, n is paper number of the author in the field, nfIt is the paper sum in the field, ciIt is the quantity to be quoted of paper i
Amount, hjIt is the H-index of author；
Author j is normalized at a distance from the f of field, facilitates and carries out follow-up data processing；Normalized formula
It is as follows：
Wherein, avg (dis (author, field)) refers to all authors at a distance from field, is asked after result of calculation adduction
Average value；
Author j is at a distance from technical ability s：
Wherein, n is the paper number of author's technical ability, nsIt is the paper sum containing the technical ability, ciIt is being drawn for paper i
With number, hjIt is the H-index of author, nisIt is the participation number of this technical ability in the paper；
Author j uses following normalization formula at a distance from technical ability s：
Wherein, avg (dis (author, skill)) refers to all authors at a distance from technical ability, is asked after result of calculation adduction
Average value；
Field f is at a distance from technical ability s：
Wherein, nfIt is the paper sum in the field, nsIt is the paper sum containing the technical ability, nfsIt is the neck containing the technical ability
Domain number；
Field f uses following normalization formula at a distance from technical ability s：
Wherein, avg (dis (field, skill)) refer to all spectra at a distance from technical ability result of calculation adduction after ask flat
Mean value；
Step 3)：By the weighing computation method on super side, the weight on super side is calculated, weight is an author in a certain neck
The certain skills proficiency in domain；According to Hypergraph Theory, the weight on super side is calculated using the deformation of gaussian kernel function, it will be in step 2)
Three distance links, obtain author a certain field certain skills horizontal parameters；
Super side right is calculated using following formula again：
Wherein, d (x, y) is the distance of author, field and technical ability between any two, and σ is the average value for distance；For i author the fields f s technical ability level value；
Step 4)：By the above process with time change, it is applied to each time to get the spy to author in a certain field
Determine changing rule of the technical ability with time change；There are various time points in the research career of author, research institution is such as replaced
With change research direction, if knowing that author in the situation of change of each technical ability of different time, can study author's technical ability at any time
Between changing rule, to find the potential rule of scientific research；To realize this target, by data set according to the growth of time,
It is annual to mark off a Sub Data Set, store this year and this year pervious data；It repeats the above steps 2) to each data subset
With step 3), the case where variation of each year technical ability of each author to author's technical ability to get changing over time is extracted in the result.
Beneficial effects of the present invention：Technical capability evaluation to author is a kind of method based on hypergraph, and this method considers opinion
The factors such as literary number and Quality of Papers, the otherness of different field, time change.The use of hypergraph concept allows this method to melt
Partner, field and technical ability, to make, process provides a kind of fine-grained evaluation schemes.Calculating author, field and skill
Can apart from when, with paper reference amount, expanded based on the Traditional measurements parameter such as H-index, ensure that reliability, simultaneously
Operation efficiency is improved using normalized, reduces error.
The present invention, which is eventually adding time factor, allows this method to analyze author field, and technical ability changes with time, after being
Continuous research provides more " raw material ".
Description of the drawings
Fig. 1 is the data prediction flow chart that the present invention carries out Plosone data sets according to requirement of experiment.
Fig. 2 is final result author technical ability radar map a of the present invention.
Fig. 3 is final result author technical ability radar map b of the present invention
Fig. 4 is the annual highest level technical ability of author incremental variations schematic diagram a at any time.
Fig. 5 is the annual highest level technical ability of author incremental variations schematic diagram b at any time.
Fig. 6 is that author's technical ability changes over time exemplary plot.
Specific implementation mode
It to make the objectives, technical solutions, and advantages of the present invention clearer, below will be to the specific embodiment party of the present invention
Formula is described in further detail.
Present invention implementation provides a kind of author's technical capability evaluation method based on hypergraph, and this method includes：
Step 1：Experimental data set of the Plosone data sets as this method is chosen, Plosone data sets are located in advance
Reason, processing procedure are as shown in Figure 1.
In order to obtain contribution of the author in paper, that is, the technical ability that author uses in this paper, present invention uses
Plosone data sets.Data set initial data is as follows：
1 Plosone data sets of table
From table 1 it follows that different technical ability quantity are very various, this may be due to not having the name of standard to technical ability
Rule causes similar technical ability to be used different expression ways.If can directly be led using the name of the original technical ability of data set
Cause result inaccurate and redundancy.
Therefore the present invention counts technical ability, and discovery has 10342 technical ability occurrence numbers to be less than 10 times, and has 21 skills
Great-jump-forward can occur to increase, therefore the casting out less than 10 times by technical ability occurrence number, increasing technical ability later with 21 great-jump-forwards is
Skill designations are sorted out on basis, finally obtain 16 skill types, each skill type is by the most technical ability of occurrence number
Title represents.
Since author's quantity is huger, there is maximum probability the problem of different authors are with same name occur, it can be to reality
It tests result and generates interference.To alleviate influence of the problem of the same name to experimental result, the present invention is according to existing differentiation algorithm of the same name, knot
The actual conditions that data set contains author's institutional affiliation are closed, the use of research institution where author's cooperation and author are reference,
Differentiation of the same name is carried out to author.The distinguishing rule of the same name that the present invention uses is：If two authors of the same name and same author had cooperation,
Then it is reasonable that the very big probability of two authors of the same name is same people.If the two same research institutions of author's subordinate of the same name, two is of the same name
The very big probability of author is same people.Since now of the same name distinguish is a research puzzle, preferable solution is there is no, the present invention is not
Differentiation problem of the same name is inquired into again.
Step 2：Three vertex on super side are combined two-by-two, are calculated later the distance between respective.
The standardization technical ability name that is obtained in step 1 and carried out the data acquisition system of differentiation of the same name, will wherein data according to
Distance calculation formula is calculated, and result is normalized later, has obtained author and field, author and technical ability, skill
At a distance from can be between field.
Step 3：By the weighing computation method on super side, the weight on this super side is calculated, this weight is an author at certain
The certain skills proficiency in one field.
According to Hypergraph Theory, super side right weight is calculated using the deformation of gaussian kernel function, calculation formula is as follows：
Wherein, d (j, s) indicate author between technical ability at a distance from, d (j, f) indicate author between field at a distance from, d (f, s)
Expression field between technical ability at a distance from, σjsIndicate the average value of all authors and technical ability distance, σjfIndicate all authors and field
The average value of distance, σfsIndicate the average value of all spectra and technical ability distance.
Show that author's technical ability is distributed for specific and concise, the technical ability point of author is embodied present invention uses radar map
Cloth.Fig. 2 and Fig. 3 gives the radar illustrated example of author's technical ability distribution, and each circle corresponds to a field in figure.
Step 4：By the above process with time change, it is applied to each time, you can obtain author in a certain field
Certain skills with time change changing rule.
The present invention being incremented by according to the time by the data of Plosone marks off a data acquisition system, marks off altogether every year
2006 to 2017 totally 12 Sub Data Sets.
Above-mentioned steps 2 and step 3 are applied to the data subset in each time, extract each author in the result not
The corresponding technical ability situation of change with the time, you can obtain author and change with time in different field difference technical ability.
The present invention is to show author's technical ability to change with time, and author's technical ability and time are integrated into line chart.Fig. 6 is provided
The line chart that author one technical ability changes over time.Since technical ability quantity is excessive, each scholar corresponds to multiple levels of skill
It changes over time, causes to be difficult to find rule, therefore author's highest level technical ability is extracted and is combined with time variable, obtain Fig. 4
With scatter plot shown in Fig. 5.
Claims (1)
1. a kind of author's technical capability evaluation method based on hypergraph, which is characterized in that steps are as follows：
Step 1)：By the field super side of composition of the technical ability and this paper of author, author in that article, the institute that statistics author participates in
There are paper, merger kinds of skill to obtain Authors of Science Articles, technical ability and the statistical data in field；The paper information that author is delivered into
Row statistics, three vertex of author, technical ability and field as super side；
Step 2)：Three vertex on super side are combined two-by-two, calculate the distance on each vertex in hypergraph；Distance between each attribute
It is calculated by following formula：
Author j is at a distance from the f of field：
Wherein, n is paper number of the author in the field, nfIt is the paper sum in the field, ciIt is the total citations of paper i, hj
It is the H-index of author；
Author j is normalized at a distance from the f of field, facilitates and carries out follow-up data processing；Normalized formula is as follows：
Wherein, avg (dis (author, field)) refers to all authors at a distance from field, is averaging after result of calculation adduction
Value；
Author j is at a distance from technical ability s：
Wherein, n is the paper number of author's technical ability, nsIt is the paper sum containing the technical ability, ciIt is that being cited for paper i is secondary
Number, hjIt is the H-index of author, nisIt is the participation number of this technical ability in the paper；
Author j uses following normalization formula at a distance from technical ability s：
Wherein, avg (dis (author, skill)) refers to all authors at a distance from technical ability, is averaging after result of calculation adduction
Value；
Field f is at a distance from technical ability s：
Wherein, nfIt is the paper sum in the field, nsIt is the paper sum containing the technical ability, nfsIt is the field number containing the technical ability；
Field f uses following normalization formula at a distance from technical ability s：
Wherein, avg (dis (field, skill)) refer to all spectra at a distance from technical ability result of calculation adduction after average；
Step 3)：By the weighing computation method on super side, the weight on super side is calculated, weight is an author in a certain field
Certain skills proficiency；According to Hypergraph Theory, the weight on super side is calculated using the deformation of gaussian kernel function, by three in step 2)
A distance link, obtain author a certain field level of skill parameter；
Super side right is calculated using following formula again：
Wherein, d (x, y) is the distance of author, field and technical ability between any two, and σ is the average value for distance；For i author the fields f s technical ability level value；
Step 4)：By the above process with time change, it is applied to each time to get the specific skill to author in a certain field
It can be with the changing rule of time change；There are various time points in the research career of author, such as replace research institution and change
Become research direction, if knowing that author in the situation of change of each technical ability of different time, just studies author's technical ability and changes over time
Rule, to find the potential rule of scientific research；It is annual to draw by data set according to the growth of time to realize this target
A Sub Data Set is separated, this year and this year pervious data are stored；2) and step it repeats the above steps to each data subset
3) the case where variation of each year technical ability of each author to author's technical ability to get changing over time, is extracted in the result.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810316651.XA CN108510205B (en) | 2018-04-08 | 2018-04-08 | Author skill evaluation method based on hypergraph |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810316651.XA CN108510205B (en) | 2018-04-08 | 2018-04-08 | Author skill evaluation method based on hypergraph |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN108510205A true CN108510205A (en) | 2018-09-07 |
| CN108510205B CN108510205B (en) | 2021-07-16 |
Family
ID=63381338
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810316651.XA Expired - Fee Related CN108510205B (en) | 2018-04-08 | 2018-04-08 | Author skill evaluation method based on hypergraph |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108510205B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112862015A (en) * | 2021-04-01 | 2021-05-28 | 北京理工大学 | Paper classification method and system based on hypergraph neural network |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102609546A (en) * | 2011-12-08 | 2012-07-25 | 清华大学 | Method and system for excavating information of academic journal paper authors |
| CN104090936A (en) * | 2014-06-27 | 2014-10-08 | 华南理工大学 | News recommendation method based on hypergraph sequencing |
| US20150347480A1 (en) * | 2014-05-30 | 2015-12-03 | Georgetown University | Process and Framework For Facilitating Data Sharing Using a Distributed Hypergraph |
| CN105956197A (en) * | 2016-06-15 | 2016-09-21 | 杭州量知数据科技有限公司 | Social media graph representation model-based social risk event extraction method |
| CN106778011A (en) * | 2016-12-29 | 2017-05-31 | 大连理工大学 | A kind of scholar's influence power appraisal procedure based on academic heterogeneous network |
| CN107273207A (en) * | 2017-05-25 | 2017-10-20 | 天津大学 | A kind of related data storage method based on hypergraph partitioning algorithm |
-
2018
- 2018-04-08 CN CN201810316651.XA patent/CN108510205B/en not_active Expired - Fee Related
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102609546A (en) * | 2011-12-08 | 2012-07-25 | 清华大学 | Method and system for excavating information of academic journal paper authors |
| US20150347480A1 (en) * | 2014-05-30 | 2015-12-03 | Georgetown University | Process and Framework For Facilitating Data Sharing Using a Distributed Hypergraph |
| CN104090936A (en) * | 2014-06-27 | 2014-10-08 | 华南理工大学 | News recommendation method based on hypergraph sequencing |
| CN105956197A (en) * | 2016-06-15 | 2016-09-21 | 杭州量知数据科技有限公司 | Social media graph representation model-based social risk event extraction method |
| CN106778011A (en) * | 2016-12-29 | 2017-05-31 | 大连理工大学 | A kind of scholar's influence power appraisal procedure based on academic heterogeneous network |
| CN107273207A (en) * | 2017-05-25 | 2017-10-20 | 天津大学 | A kind of related data storage method based on hypergraph partitioning algorithm |
Non-Patent Citations (1)
| Title |
|---|
| JIAJUN BU ET AL.: "Music Recommendation by Unified Hypergraph:Combining Social Media Information and Music Content", 《PROCEEDINGS OF THE 18TH ACM INTERNATIONAL CONFERENCE ON MULTIMEDIA》 * |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112862015A (en) * | 2021-04-01 | 2021-05-28 | 北京理工大学 | Paper classification method and system based on hypergraph neural network |
Also Published As
| Publication number | Publication date |
|---|---|
| CN108510205B (en) | 2021-07-16 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Wang et al. | Correlation structure and dynamics of international real estate securities markets: A network perspective | |
| Tsen | THE DETERMINANTS OF FOREIGN DIRECT INVESTMENT IN THE MANUFACTURING INDUSTRY OF MALAYSIA. | |
| Zebende et al. | Detrended multiple cross-correlation coefficient | |
| CN109408712A (en) | A kind of construction method of travel agency user multidimensional information portrait | |
| Coulombe et al. | Economic growth and regional income disparities in Canada and the Northern United States | |
| Çiçek et al. | The asymmetry of inflation adjustment in Turkey | |
| Usman et al. | Effects of capital flight and Its macroeconomic determinants in agricultural growth in Nigeria (1970-2013) | |
| Ji et al. | Dependence structure between China’s stock market and other major stock markets before and after the 2008 financial crisis | |
| Chen et al. | Analyzing the co-movement and its spatial–temporal patterns in Chinese stock market | |
| CN108510205A (en) | A kind of author's technical capability evaluation method based on hypergraph | |
| Fungáčová et al. | Determinants of financial inclusion in Asia | |
| CN103955715B (en) | Automatic person identification method based on barefoot or stockinged footprint image | |
| Queiroz | Estimating maternal mortality differentials using census data: experience in Honduras | |
| Link et al. | The influence of the quantity and quality of education on black-white earnings differentials: Some new evidence | |
| Siliverstovs et al. | Recessions as breadwinner for forecasters state-dependent evaluation of predictive ability: Evidence from big macroeconomic us data | |
| Feldkircher et al. | The Transmission of Euro Area Interest Rate Shocks to Asia--Do Effects Differ When Nominal Interest Rates are Negative? | |
| Afzali | The Impact of Political Stability on Foreign Direct Investment in South Asian Countries Focusing on Afghanistan | |
| Zulkifli et al. | Portfolio diversification: the role of information technology in future investment decision-making | |
| Kaur et al. | Factual dimension identification and usage for real estate framework | |
| Liumei et al. | Analysis of system implementation effect based on Bayes analysis of imbalanced measures | |
| ADEREMI et al. | DETERMINANTS OF FOREIGN DIRECT INVESTMENT INFLOWS: EVIDENCE FROM A PANEL ANALYSIS OF SOUTH ASIAN COUNTRIES | |
| Li et al. | Research on structural correlation of HS 300 stock index based on AR (n)-XARCH-Copula model | |
| Kishor et al. | Business cycle synchronization in the CIS region | |
| Wang | A study on the influencing factors of CPI bias in different groups in Canada | |
| Anderson | Community chest campaigns as an index of community integration |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20210716 |