Scientometric Indicators and Machine Learning-Based Models for Predicting Rising Stars in Academia

摘要

Newly recruited researchers who are expected to outstandingly surpass their peers in the quality of their work, are often considered as substantial assets in universities and research & development entities. Foreseeably identifying such Rising Stars is vital for highly competitive and profitable institutes and organizations. In this paper, we propose models based on a set of Scientometric Indicators to predict rising stars in academia. In addition, we define the rising stars problem in a comprehensive and methodological manner. Machine learning techniques are applied on actual data subsets collected from the Web of Science (WoS) data source. Our experimental results show that the proposed models and indicators can be used effectively in predicting future rising stars.