A Study On The Prediction Of Academic Influence And Academic Innovation Of Chinese Sports Scholars Based On Graph Neural Network

As research collaboration becomes a common phenomenon in scientific research,the study of the influence of authors’ academic influence and academic innovation has gradually become an important issue in the field of research collaboration.As we all know,academic influence and academic innovation is an important basis for assessing the academic level of authors.The higher the academic influence and academic innovation of an author,the stronger his or her academic research ability and the greater the influence of his or her academic results,which play a great role in promoting the development of the disciplinary field.By evaluating the academic influence and academic innovation of authors,not only can we establish the correct value orientation of scientific research work and promote the sustainable and stable development of scientific research work,but also provide an objective basis for teacher recruitment,project application and government funding.By studying how research collaboration affects authors’ research performance,we can identify the factors that affect authors’ research performance and their intrinsic correlation,so as to optimise the way of research collaboration and improve authors’ research performance and level.Based on relevant studies at home and abroad,this study uses journal paper data from two Chinese databases,CSSCI and Peking University’s Chinese Core Journals Catalogue,to construct an academic collaboration network of authors in the field of sports science in China,and quantitatively analyse the relationship between academic collaboration networks on authors’ academic influence and academic innovation,guided by citation network analysis and the law of document concentration and dispersion.Finally,the graph convolutional neural network algorithm was used to carry out a prediction study on the academic influence and academic innovativeness of authors.The following conclusions were drawn from this study: 1.the trend of collaboration among authors in the field of kinesiology in China is becoming more and more obvious,with the proportion of co-authored articles reaching as high as 80% in 2021 compared to less than 40% in 1990;2.the overall academic influence and academic innovation of authors in the field of kinesiology in China is at a medium level,with very few top sports authors having high academic influence and academic innovation 3.there is a significant correlation between the academic influence and academic innovation of authors in the field of kinesiology and the attributes of network structure and network nodes in the academic cooperation network of authors in China,and academic influence shows a significant positive correlation with academic innovation and degree centrality and structural holes in the academic cooperation network,and a negative correlation with the clustering coefficient;4.the academic influence and academic innovation of authors in the field of kinesiology in China constructed by using the graph convolutional neural network algorithm The prediction model of academic influence and academic innovation of authors in the field of kinesiology in China has the characteristics and advantages of high accuracy and good validity,and the effect of its final prediction results is much better than that of the linear regression equation model.This study introduces the graph neural network algorithm into the prediction system of academic influence and academic innovation of authors in the field of physical education in China,which to a certain extent promotes the interdisciplinary field of physical education and computer science,and can effectively find potential authors.However,there is still room for improvement in this study: 1.to build a more effective and scientific evaluation system for academic influence and academic innovation;2.to improve data collection and analysis capabilities and obtain more relevant data such as subject data and conference data;3.to pay more attention to the prediction research and analysis of individual authors in the field of kinesiology in China;4.to obtain more personal information about authors in the field of kinesiology in China to improve;The study only investigates the academic influence and academic innovation of authors in the field of kinesiology in China,whether the conclusions drawn in this study can be extended to other fields,and whether the conclusions have continuity.