How Can I Grow!!: Analysis of Co-authorship Network of Early Career Stage Researchers

Identifying and providing assistance to young researchers in the early stages of their careers is crucial to foster their development and promote the advancement of scientific knowledge within their specific fields. Within this document, we introduce a machine learning-driven approach designed to pinpoint ECRs within extensive bibliographic datasets. Our suggested framework capitalizes on the set of features selected in a dataset, encompassing details like publication history, citations, academic associations, and collaborative author networks. In our models, we make use of the comprehensive and varied AMiner Author dataset. By employing a novel approach, we have achieved a notable level of accuracy in distinguishing ECRs, effectively separating them from established researchers, with “h-index” distributions as a key discriminator.