On the Discovery of Success Trajectories of Authors

TL;DR

This paper identifies six distinct success trajectories of authors in computer science and physics, and develops a model to predict future success based on early career data, outperforming baselines.

Contribution

It introduces a novel classification of author success trajectories and a predictive model for early career success in scientific research.

Findings

Six success trajectory types identified in authors' careers.

The stratification model predicts future success with 15.68% improvement.

Model outperforms baseline methods in accuracy.

Abstract

Understanding the qualitative patterns of research endeavor of scientific authors in terms of publication count and their impact (citation) is important in order to quantify success trajectories. Here, we examine the career profile of authors in computer science and physics domains and discover at least six different success trajectories in terms of normalized citation count in longitudinal scale. Initial observations of individual trajectories lead us to characterize the authors in each category. We further leverage this trajectory information to build a two-stage stratification model to predict future success of an author at the early stage of her career. Our model outperforms the baseline with an average improvement of 15.68% for both the datasets.