CN109614502A - A method of R&D institution's influence power is assessed based on academic big data - Google Patents

A method of R&D institution's influence power is assessed based on academic big data Download PDF

Info

Publication number
CN109614502A
CN109614502A CN201811414500.4A CN201811414500A CN109614502A CN 109614502 A CN109614502 A CN 109614502A CN 201811414500 A CN201811414500 A CN 201811414500A CN 109614502 A CN109614502 A CN 109614502A
Authority
CN
China
Prior art keywords
institution
paper
studied
academic
data
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
Application number
CN201811414500.4A
Other languages
Chinese (zh)
Inventor
刘建国
王江盼
朱熹华
李超然
郭强
江明珠
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Shanghai university of finance and economics
Original Assignee
Shanghai university of finance and economics
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Shanghai university of finance and economics filed Critical Shanghai university of finance and economics
Priority to CN201811414500.4A priority Critical patent/CN109614502A/en
Publication of CN109614502A publication Critical patent/CN109614502A/en
Pending legal-status Critical Current

Links

Classifications

    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06QINFORMATION 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
    • G06Q10/00Administration; Management
    • G06Q10/06Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
    • G06Q10/063Operations research, analysis or management
    • G06Q10/0639Performance analysis of employees; Performance analysis of enterprise or organisation operations
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06QINFORMATION 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
    • G06Q10/00Administration; Management
    • G06Q10/10Office automation; Time management
    • G06Q10/101Collaborative creation, e.g. joint development of products or services

Landscapes

  • Business, Economics & Management (AREA)
  • Human Resources & Organizations (AREA)
  • Engineering & Computer Science (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Strategic Management (AREA)
  • Economics (AREA)
  • General Business, Economics & Management (AREA)
  • Marketing (AREA)
  • Operations Research (AREA)
  • Quality & Reliability (AREA)
  • Tourism & Hospitality (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Development Economics (AREA)
  • Educational Administration (AREA)
  • Data Mining & Analysis (AREA)
  • Game Theory and Decision Science (AREA)
  • Information Retrieval, Db Structures And Fs Structures Therefor (AREA)

Abstract

The invention discloses a kind of methods assessed based on academic big data R&D institution's influence power, the reference data of all R&D institutions involved in main paper data and paper using in academic paper database and paper, that establishes dependence R&D institution draws directed networks altogether, with the importance of paper drawn weight altogether and measure paper, and the academic for the R&D institution studied using the method quantitative assessment of R&D institution's signature location information building weighting coefficient function.The present invention establishes directed networks according to academic paper reference data, the influence value of each unit is calculated for the method that the paper of multiple unit cooperations uses weighting coefficient, the academic of each R&D institution is scientificlly and effectively evaluated, to provide a kind of effective and reasonable R&D institution's academic evaluation method.

Description

A method of R&D institution's influence power is assessed based on academic big data
Technical field
The present invention relates to technical field of data processing, and in particular to based on academic big data to scientific research in sphere of learning In the case of the method more particularly to scientific research that unit influence power is assessed are collaborateed, a kind of section based on weighting coefficient directed networks Grind the evaluation method of unit academic.
Background technique
Today's society, R&D institution are the important participants of scientific research, have made major contribution for academic research.However There is the different capacities of scientific research in different R&D institutions, and the scientific research level of R&D institution reflects the academic shadow an of R&D institution Ring power.The quantized result of academic provides important for the examination & approval of state research project, the attraction of the talent, ranking of mechanism etc. Foundation.
Furthermore with the complexity of social concern, cooperate between R&D institution it is more and more common, between R&D institution cooperation become To day-to-day, the assignment problem of scientific achievement is just more and more obvious.Currently, existing science allocation algorithm is mainly according to being drawn The frequency measures paper importance, and mean allocation article is worth to collaborateing main body.However whether an article is quoted by many Factor influences, while the method for salary distribution divided equally used, reasonability still remain to be discussed.
Summary of the invention
In view of the deficiencies of the prior art, the present invention is intended to provide it is a kind of based on academic big data to R&D institution's influence power The method assessed draws directed networks between building paper, with being total to for paper by network structure and R&D institution's information altogether Draw the times cited that weight replaces paper, the significance level for measuring paper is gone from R&D institution's angle, while considering R&D institution Signature order factor in paper carries out accurate evaluation to R&D institution's academic in the case of collaboration.
To achieve the goals above, The technical solution adopted by the invention is as follows:
A method of R&D institution's influence power is assessed based on academic big data, is included the following steps:
S1, acquisition R&D institution's all academic papers delivered and its data being cited to be studied, remember to be studied R&D institution is Ii
S2, the data collected in step S1 are set up according to adduction relationship and are drawn data set P={ p1,p2,...,pm}、 It applies and draws data set C={ c1,c2,...,clAnd directed networks;Drawn data set and is owned for what R&D institution to be studied delivered The set of academic paper, applies that draw data set be at least one set for applying introduction text referred to by argument according to concentration, oriented Comprising being drawn data set, applying the relationship drawing data set and reference between the two and being cited in network;
S3, to being drawn data set P={ p1,p2,...,pmIn each paper proceed as follows:
S3.1, note pjIn have a signature mechanism, signing includes R&D institution I to be studied in mechanismi, j=1,2 ..., m;From C={ c1,c2,...,clFind out reference paper pjApply introduction collected works close Cj={ c1,c2,...,cb, and from P={ p1, p2,...,pmIn filter out by introduction collected works close Dj={ d1,...,dc};DjIn include paper and meanwhile meet two conditions: the First is that at least with paper pjThere is an identical signature mechanism, second is simultaneously by CjAt least one of apply introduction text reference;
S3.2, R&D institution I to be studied is calculated according to the following formulaiIn DjThe contribution rate in h papers in setWith R&D institution I to be studiediIn DjIt is affixed one's name in the signature sequence r and h papers in h papers in set The total k of name mechanism is related:
S3.3, D is calculatedjH papers in set are aggregated CjIn paper reference number wj,h
S3.4, R&D institution I to be studied is calculated as followsiFrom paper pjThe scientific research contribution amount of middle acquisition
R&D institution I to be studied will be obtained after the completion of S4, step S3iThe scientific research contribution obtained in m different papers AmountIt is made into accumulation calculating as the following formula, obtains R&D institution I to be studiediScientific research influence power
Further, when the scientific research influence power to the R&D institution to be required study while when calculating, in step S1 All academic papers that all academic papers and its data being cited are delivered for the R&D institution that is required study and Its data being cited, all academic papers delivered by the R&D institution that data set is required study by finger is drawn in step S2 Set, for each paper p in step S3j, while calculating paper pjSignature mechanism in include the section to be required study Unit is ground from paper pjThe scientific research contribution amount of middle acquisition
Further, when the scientific research influence power to the R&D institution to be required study is respectively calculated, in step S1 All academic papers and its data being cited are delivered all for the targeted R&D institution to be studied of this time calculating Academic paper and its data being cited, the data set that drawn in step S2 is the scientific research to be studied for referring to that this time calculating is targeted The set for all academic papers that unit is delivered.
The beneficial effects of the present invention are: the present invention is based on weighting coefficient directed networks to comment R&D institution's influence power Estimate, by network structure and R&D institution's information, draw directed networks altogether between building paper, replaces opinion with the weight of drawing altogether of paper The times cited of text goes the significance level for measuring paper from R&D institution's angle, while considering administration of the R&D institution in paper Name order factor carries out accurate evaluation to R&D institution's academic in the case of collaboration.
Detailed description of the invention
Fig. 1 is the General Implementing flow diagram of the embodiment of the present invention 2;
Fig. 2 is that physics periodical APS data set is utilized in the embodiment of the present invention 3, using random edged p1Addition noise Method to R&D institution's academic discriminating power result figure;
Fig. 3 is that physics periodical APS data set is utilized in the embodiment of the present invention 3, using random cut edge reconnection p2Addition The method of noise is to R&D institution's academic discriminating power result figure.
Specific embodiment
Below with reference to attached drawing, the invention will be further described, it should be noted that the present embodiment is with this technology side Premised on case, the detailed implementation method and specific operation process are given, but protection scope of the present invention is not limited to this reality Apply example.
Embodiment 1
A method of R&D institution's influence power is assessed based on academic big data, is included the following steps:
S1, acquisition R&D institution's all academic papers delivered and its data being cited to be studied, remember to be studied R&D institution is Ii
S2, the data collected in step S1 are set up according to adduction relationship and are drawn data set P={ p1,p2,...,pm}、 It applies and draws data set C={ c1,c2,...,clAnd directed networks;Drawn data set and is owned for what R&D institution to be studied delivered The set of academic paper, applies that draw data set be at least one set for applying introduction text referred to by argument according to concentration, oriented Comprising being drawn data set, applying the relationship drawing data set and reference between the two and being cited in network;
S3, from being drawn data set P={ p1,p2,...,pmOne paper of middle selection, it is denoted as pj
S4, note pjIn have a signature mechanism, including R&D institution I to be studiedi;From C={ c1,c2,...,clFind out and draw With paper pjApply introduction collected works close Cj={ c1,c2,...,cb, and from P={ p1,p2,...,pmIn filter out by introduction collected works Close Dj={ d1,...,dc};DjIn include paper and meanwhile meet two conditions: first be at least with paper pjHave one it is identical Mechanism is signed, second is simultaneously by CjAt least one of apply introduction text reference;
S5, R&D institution I to be studied is calculated according to the following formulaiIn DjThe contribution rate in h papers in set With R&D institution I to be studiediIn DjMechanism is signed in the signature sequence r and h papers in h papers in set Total k it is related:
S6, D is calculatedjH papers in set are aggregated CjIn paper reference number wj,h(i.e. set CjIn have it is more Few piece paper refers to DjH papers in set);
S7, R&D institution I to be studied is calculated as followsiFrom paper pjThe scientific research contribution amount of middle acquisition
It is also right in the present embodimentIt is standardized:
The method that standardization uses is shown below,Value after standardization is For h papers In all R&D institutionsSummation;
S8, to P={ p1,p2,...,pmIn all paper carry out the processing of step S4-S7, obtain scientific research to be studied Unit IiThe scientific research contribution amount obtained in m different papers, then makees accumulation calculating as the following formula, obtains scientific research to be studied Unit IiScientific research influence power
To the R&D institution to be required study simultaneously or separately according to the method for above-mentioned steps S1-S8, institute can be obtained There is the value of the scientific research influence power of R&D institution and carries out ranking.
Embodiment 2
As shown in Figure 1, a kind of commenting R&D institution's influence power based on academic big data provided by the present embodiment The method estimated is to calculate simultaneously the value of the scientific research influence power of the R&D institution to be required study, includes the following steps:
One, R&D institution's all academic papers delivered and its data being cited to be studied, section to be studied are acquired Grinding unit includes I1、I2、I3、I4
Two, the data collected in step 1 are set up according to adduction relationship and is drawn data set, applies and draw data set, and established Directed networks.In the present embodiment, being drawn data set is P={ p1,p2,p3,p4, it applies and draws data set for C={ c1,c2,c3,c4, c5, shown in the left side first box such as Fig. 1 for the directed networks established.
Three, paper p is selected according to concentration from by argument1
Four, paper p1In signature mechanism include R&D institution I to be studied1And I2.From C={ c1,c2,c3,c4,c5In Find out reference paper p1Apply introduction collected works close C1={ c1,c3, after filter out by introduction collected works close D1={ d1,d2,d3,d4, such as Shown in Fig. 1.
Five, R&D institution I to be studied is calculated1And I2In D1Contribution rate in each piece paper of setThis and R&D institution I to be studied1And I2In D1 Signature position in each piece paper of set is related.
The calculating process of each contribution rate are as follows:
Similarly obtain
In the present embodiment, for ease of calculation, by R&D institution I to be studied1And I2In D1In each piece paper of set Contribution rate To contribute allocation matrix A1Form table Show:
Six, D is calculated1The reference amount w of h papers in set1,h, it is defined as D1H papers in set are aggregated C1In paper reference number, i.e. set C1In how many paper refer to D1H papers in set.
In the present embodiment, paper p1By paper c1And c3Reference, therefore paper p1Draw weight w altogether with itself1,1=2; Paper p2Only by paper c1Reference, therefore paper p2With paper p1Draw weight w altogether1,2=1.Similarly, available paper p3And opinion Literary p1Draw altogether weight be 2, paper p4With paper p1Draw altogether weight be 1.In the present embodiment, using drawing weight matrix w altogether1 It indicates:
w1=[2,1,2,1].
Seven, to set D1In each paper calculateWithHerein Using scientific research influence power matrix M1It indicates:
It is obtained after being standardized
Eight, according to the method for step 4-seven, aforesaid operations are carried out to each paper in P, obtaining scientific research influences torque Battle array M2、M3、M4:
The scientific research influence power matrix M that will be calculated1、M2、M3、M4It is added, available scientific research contribution collection M=[I1, I2,I3,I4]=[0.36,1.73,0.55,1.36].
Nine, it contributes collection M to arrange in descending order scientific research, obtains the ranking vector R=[I of R&D institution to be studied2,I4,I3, I1]。
It should be noted that the present embodiment method is to calculate the scientific research influence power of each R&D institution to be studied simultaneously, But each R&D institution to be studied individually can also be calculated according to the method for step 1 to eight respectively, final basis Calculated result carries out ranking.
Embodiment 3
The present embodiment passes through the performance of experimental verification the method for the present invention.
The adduction relationship and the corresponding R&D institution's information of every paper of paper are extracted, from APS data set with this shape At paper citation network.
Wherein, the data selected in the present embodiment in APS data set from 1893 to 2009 year, screening are contained 18987 mechanisms, 443217 articles, 4710547 reference records.
It is tested by Kendall's coefficient curve to effect of the method for the present invention in data set, by original quotation Man made noise's data (random edged p is added in data set1With random cut edge reconnection p2), scientific research list is obtained by the method for embodiment 1 Position academic score carries out the sequence of influence power size to R&D institution, by learning with the R&D institution not plus before noise data The coincidence degree of art influence power ranking result evaluates the method for the present invention result.The result of APS data set respectively such as Fig. 2 and Shown in Fig. 3, as seen from the figure, in the experiment for carrying out random edged and random cut edge reconnection, the method for the present embodiment proposition In p1、p2When taking different value, Kendall's coefficient value is above the algorithm that TC algorithm and Shen Huawei are proposed.
For those skilled in the art, it can be made various corresponding according to above technical solution and design Change and modification, and all these change and modification should be construed as being included within the scope of protection of the claims of the present invention.

Claims (3)

1. a kind of method assessed based on academic big data R&D institution's influence power, which is characterized in that including walking as follows It is rapid:
S1, acquisition R&D institution's all academic papers delivered and its data being cited to be studied, remember scientific research to be studied Unit is Ii
S2, the data collected in step S1 are set up according to adduction relationship and are drawn data set P={ p1,p2,...,pm, Shi Yin Data set C={ c1,c2,...,clAnd directed networks;Being drawn data set is all science that R&D institution to be studied delivers The set of paper, applies that draw data set be at least one set for applying introduction text referred to by argument according to concentration, directed networks In comprising being drawn data set, applying the relationship drawing data set and reference between the two and being cited;
S3, to being drawn data set P={ p1,p2,...,pmIn each paper proceed as follows:
S3.1, note pjIn have a signature mechanism, signing includes R&D institution I to be studied in mechanismi, j=1,2 ..., m;From C ={ c1,c2,...,clFind out reference paper pjApply introduction collected works close Cj={ c1,c2,...,cb, and from P={ p1, p2,...,pmIn filter out by introduction collected works close Dj={ d1,...,dc};DjIn include paper and meanwhile meet two conditions: the First is that at least with paper pjThere is an identical signature mechanism, second is simultaneously by CjAt least one of apply introduction text reference;
S3.2, R&D institution I to be studied is calculated according to the following formulaiIn DjThe contribution rate in h papers in set With R&D institution I to be studiediIn DjThe total of mechanism is signed in the signature sequence r and h papers in h papers in set Number k is related:
S3.3, D is calculatedjH papers in set are aggregated CjIn paper reference number wj,h
S3.4, R&D institution I to be studied is calculated as followsiFrom paper pjThe scientific research contribution amount of middle acquisition
R&D institution I to be studied will be obtained after the completion of S4, step S3iThe scientific research contribution amount obtained in m different papersIt is made into accumulation calculating as the following formula, obtains R&D institution I to be studiediScientific research influence power
2. the method according to claim 1, wherein when the scientific research influence power to the R&D institution to be required study Simultaneously when being calculated, all academic papers described in step S1 and its data being cited are the R&D institution that is required study All academic papers delivered and its data being cited, the scientific research to be required study by data set is drawn by finger in step S2 The set for all academic papers that unit is delivered, for each paper p in step S3j, while calculating paper pjSignature mechanism In include the R&D institution to be required study from paper pjThe scientific research contribution amount of middle acquisition
3. the method according to claim 1, wherein when the scientific research influence power to the R&D institution to be required study When being respectively calculated, all academic papers described in step S1 and its data being cited are that this time calculating is targeted wait grind All academic papers and its data being cited that the R&D institution studied carefully delivers are drawn data set to refer to the secondary meter in step S2 The set for all academic papers that targeted R&D institution to be studied delivers.
CN201811414500.4A 2018-11-26 2018-11-26 A method of R&D institution's influence power is assessed based on academic big data Pending CN109614502A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201811414500.4A CN109614502A (en) 2018-11-26 2018-11-26 A method of R&D institution's influence power is assessed based on academic big data

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201811414500.4A CN109614502A (en) 2018-11-26 2018-11-26 A method of R&D institution's influence power is assessed based on academic big data

Publications (1)

Publication Number Publication Date
CN109614502A true CN109614502A (en) 2019-04-12

Family

ID=66004940

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201811414500.4A Pending CN109614502A (en) 2018-11-26 2018-11-26 A method of R&D institution's influence power is assessed based on academic big data

Country Status (1)

Country Link
CN (1) CN109614502A (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111831905A (en) * 2020-06-19 2020-10-27 中国科学院计算机网络信息中心 Recommendation method and device based on team scientific research influence and sustainability modeling

Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106650276A (en) * 2016-12-29 2017-05-10 大连理工大学 Scholar impact evaluation method based on academic big data
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

Patent Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106650276A (en) * 2016-12-29 2017-05-10 大连理工大学 Scholar impact evaluation method based on academic big data
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

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
崔景洋: "一种基于PageRank和时间衰退因子的作者影响力评价方法", 《廊坊师范学院学报(自然科学版)》 *
谢瑞霞等: "基于加权被引频次与署名顺序的作者影响力评价指标构建", 《情报科学》 *

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111831905A (en) * 2020-06-19 2020-10-27 中国科学院计算机网络信息中心 Recommendation method and device based on team scientific research influence and sustainability modeling
CN111831905B (en) * 2020-06-19 2023-06-06 中国科学院计算机网络信息中心 Recommendation method and device based on team scientific research influence and sustainability modeling

Similar Documents

Publication Publication Date Title
Lee et al. Comparative evaluation of performance of national R&D programs with heterogeneous objectives: A DEA approach
CN107784426A (en) Post distribution method, device and the equipment of a kind of employee
Duan et al. Comparison of two coronal magnetic field models to reconstruct a sigmoidal solar active region with coronal loops
Giscard et al. Evaluating balance on social networks from their simple cycles
Bovolenta et al. Binding energy evaluation platform: A database of quantum chemical binding energy distributions for the astrochemical community
Leydesdorff et al. Are university rankings statistically significant? A comparison among Chinese universities and with the USA
CN103679291A (en) Patent value assessment method
CN107944750A (en) A kind of poverty depth analysis method and system
CN109614502A (en) A method of R&D institution's influence power is assessed based on academic big data
CN110750572A (en) Adaptive method and device for heuristic evaluation of scientific and technological achievements
CN106022975A (en) Water resource development and utilization efficiency evaluation method and application thereof
CN104376042B (en) A kind of item database examination question acquisition methods and system
Zhang Blunt to sharpened razor: incremental reform and distortions in the product and capital markets in China
Okoń-Horodyńska et al. Gender in the creation of intellectual property of the selected European Union countries
Benedetti et al. Capturing the phase diagram of (2+ 1)-dimensional CDT using a balls-in-boxes model
Pokushko et al. Application of data envelopment analysis method for assessment of performance of enterprises in fuel and energy complex
CN109325675A (en) A kind of capital electrical index calculation method and system
CN109146259A (en) A kind of independent sale of electricity development of company Assessment Method on Potential based on grey Cluster Evaluation
Davidovsky System Analysis and Forecasting of the Relationship Between Economic Factors and the Epidemic Process COVID-19 to Optimize International Transport Communications in the Context of a Pandemic
CN109215793A (en) A kind of financial industry user experience health degree assessment system and its application method
Aderemi et al. What Drives Foreign Direct Investment Inflows? Evidence from a Panel Analysis of BRICS Countries
Hall Adding Meaning to Measurement: Evaluating Trends and Differences in Innovation Capacity Among the States
Banihashemi et al. Cross efficiency evaluation with negative data in selecting the best of portfolio using OWA operator weights
Wang et al. Research on the Comprehensive Evaluation of Graduate Innovation Ability in Various Regions of China
Chen et al. The Impact of PPP Mode on Poverty Alleviation: A Difference-in-Difference Analysis

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: 20190412