Abstract
AbstractThis paper selects “Nature” papers as data sources and introduces supplementary evaluation indexes to make up for the deficiencies of traditional citation evaluation. Machine learning analysis methods such as correlation analysis, factor analysis, and principal component analysis were used to analyze the index data, and a comprehensive influence evaluation system was constructed. The results show that the comprehensive evaluation model is more reasonable and comprehensive than the traditional evaluation. This model can calculate the comprehensive influence scores of the papers so that the papers can be ranked more intuitively. By analyzing the word frequency of keywords in papers with a high comprehensive evaluation, the focus of future research can be predicted to a certain extent. This paper provides a reference for experts and scholars to select scientific research topics and scientific decisions and also provides new ideas for the selection of comprehensive influence evaluation indexes and optimization of evaluation models for academic papers.KeywordsComprehensive influenceHot papersBibliometricAltmetrics
Published Version
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call