Scholar-Institution Matching Based On Academic Profiling

Nowadays,the rapid development of network science theory and data analysis methods and technologies has made academic data with exponential growth easier to be mined and statistically analyzed,and it has also attracted a large number of researchers to conduct related research on scholarly big data.In scientific research field,all scholars will choose at least one institution as a research unit in their academic careers.Academic institutions also need to inject fresh blood to promote their scientific research development.This actual demand is related to the vital interests of each scholar and institution.This paper proposes a preference list suitable for each other based on academic profiling and matches them according to certain rules.According to empirical research,the main contributions of this article are as follows:(1)Evaluation of academic potential based on network location.Based on the three fields of computer science,biology,and psychology,this article proposes a indicator to evaluate and quantify the academic potential of scholars.By analyzing the citation relationship between scholars in the scholar citation network and their existing academic achievements,an algorithm is proposed to calculate the scholar's current academic potential.Besides,this paper also compares and analyzes the existing indicators,showing the advantages,characteristics,and differences of the proposed indicator with other indicators int terms of extensive experiments.(2)Analysis of research complexity based on the citation of the paper.By setting the "World University Academic Ranking" proposed by Shanghai Jiao Tong University as the benchmark,this article uses the top 100 universities as an experimental object,and transplants the indicator of "economic complexity" in economics into the academic field.By introducing the concept of "research complexity" and optimizing the existing algorithms,this paper also applies this concept at the national level.By using the filtered data in 20 developed countries and 20 developing countries,the research complexity of the target institution and country are calculated.Finally,the characteristics of the indicator and correlation analysis with other relevant indicators are analyzed.(3)Scholar-institution matching based on academic profiling.Academic profiling is an important branch of scholarly big data research.This article embeds the above two indicators as part of academic profiling into the algorithm of scholar-institution matching.By considering the research direction,academic research level,and other impact factors of scholars and institutions,this paper calculates the similarity between scholars and institutions and propose the preference lists of scholars and institutions.After proving the effectiveness of the preference list,this article matches both sides according to the predefined conditions.