Research And Application Of Scientific Team Performance Evaluation Based On Mult-source Information

In the current era of knowledge economy, the scientific research team takes the knowledge sharing, the professional complementary, the information exchange as the organization pattern, integrate superior resources, exchange innovative ideas and achieve common goals. It shows great advantages for technological progresses, research projects, academic researches and invention patents. Therefore, it is of great value and practical significance to evaluate the performance of scientific research team. Researchers should effectively understand the potential factors affecting the performance of the team as well as the corresponding influence weight, in order to develop and promote scientific research.We applied social network analysis, text mining methods and SPSS statistical analysis, combining with traditional bibliometric to develop a comprehensive study on the performance evaluation of scientific research team and put forward a fair performance evaluation system based on factual data. It provided a new idea for the research on the performance evaluation. We took the cooperative relationship of scientific researchers in the team as the breakthrough point, used the theory of social network analysis and network indicators to find the implicit index to evaluate team’s performance from the perspective of cooperation network, cohesive subgroups and the core members of the team. Text mining method was used to obtain the main research points and their distribution. Then we dug out the factors affecting the performance evaluation based on the diversity,compatibility and complementarity of research orientation of the team. We used the bibliometric method to have more comprehensive evaluation indexes which can describe the actual research output of the team and scientific research level of the researchers. In the empirical analysis part, we used SPSS single variance analysis method to analyze the correlation between indexes in these three categories of indicators and determined the index group consisting of indexes which are pairwise independent. The SPSS multivariate regression analysis method was applied to determine the weight of their influence for the team performance. Finally, we put forward a relatively objective, fair and reasonable performance evaluation system of scientific research team and verified its effectiveness.