An Analysis of the Relationship Between Author’ s Academic Age and Scientific Collaboration Using Multivariate Mathematical Regression

This article is based on the Microsoft Academic Graph (MAG) database and selects publication and citation data in the disciplines of economics and chemistry from 1970 to 2020. The academic age of each scholar is dynamically defined, and the cluster analysis method is used to classify the academic age of the author, namely: young scholars, middle-aged scholars, and senior scholars. The authors’ cooperation is divided using the maximum and minimum method. Based on existing data, a negative exponential function model is constructed to investigate the relationship between the number of publications and the type of author collaboration. The fitted parameters are statistically tested using mathematical regression. On this basis, the relationship between academic age and scientific research output is analyzed. It is concluded that collaborating with young scholars contributes more to scientific research output in terms of publication count, but there are disciplinary differences in citation count: for economics, authors of other ages groups contribute more when collaborating with middle-aged scholars; in chemistry, authors of other age groups contribute more when collaborating with young scholars. An average academic age at peak contribution under optimal collaboration scale is also given.The abstract needs to summarize the content of the paper.