期刊
SCIENTOMETRICS
卷 127, 期 4, 页码 1867-1882出版社
SPRINGER
DOI: 10.1007/s11192-022-04286-w
关键词
Citation counts; H-index; Deep learning; Scientific impact; Prediction
资金
- National Natural Science Foundation of China [71731002]
- China Scholarship Council (CSC)
Predicting the future career of individual scientists is crucial for recruitment, promotion, and grant management in scientific research. Existing studies primarily focus on macro level performance, while this research tackles the micro level tasks of predicting impact and publication dates of future papers. A deep learning method is proposed and outperforms the state-of-the-art in this area.
Predicting the future career of individual scientists is an important yet challenging problem with numerous applications such as recruitment of scientific research positions, promoting outstanding academic staff, and managing scientific grant proposals. Despite that much effort has been devoted to predict scientists' future performance and success, yet these works focus on the macro future performance of scholars from the perspective of their career ages. A related but different task is to predict the impact and publication date of each future paper. We regard this micro level prediction problem as a dynamic series auto-regression task, and a deep learning method is designed to solve it. The experiments show that our method outperforms the state-of-the-art method in this issue.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据