Research On The Innovation Evaluation Of Academic Papers Based On Text Mining

With the development of the open access movement,the full text of academic papers,reviewers’ opinions and other materials are more and more easily obtained.These texts contain a lot of useful information,which can provide new perspectives and methods for evaluating the innovation of academic papers.At the same time,with the continuous development of computer technology,natural language processing technology has gradually been applied to many research fields other than the computer field.In the field of evaluative metrology,natural language processing technology has gradually become a research hotspot.With the help of text mining technology,this paper excavates the content of innovation from the reviewer’s comment text,and uses the machine learning model and "voting principle" to construct an evaluation index of academic paper innovation based on the sentiment analysis of the innovation of the review text.Then,the constructed index is tested for rationality by combining the subject innovation measure value of the academic paper itself and the total score of the paper on the Open Review website.Finally,an academic innovation evaluation index system integrating the above construction indicators is constructed,and the entropy method is used to weight the indicators,and comprehensively evaluate the innovation of academic papers.The research results of this paper transform the reviewers’ unstructured and nonquantitative comment texts into quantitative innovation evaluation indicators,which can more intuitively show the innovation of academic papers to the public,and improve the innovation evaluation indicators of academic papers.The system construction provides a reference,which can help realize the fairness,scientificity and timeliness of the evaluation of the innovation of academic papers,improve the utilization rate of academic papers with high innovation,enhance the transformation ability of scientific and technological achievements,and help the implementation of the national innovation-driven development strategy.