The Risks of Introspection: A Quantitative Analysis of Influence Between Scientific Communities

Abstract

Research impact is increasingly evaluated by quantitative analysis of citations, such as the h-index or conference impact factor. However, little attention has been given to understanding the patterns of influence and dependence that exist between groups of scientific communities. We propose that an analysis of inter-community influence offers valuable insights into how scientific communities evolve in terms of growth, stability, decline, information exchange, and impact. We present a computational model for inter-community influence in science and evaluate it on 19 years of data from communities in the field of Artificial Intelligence (AI). We uncover and explain the dynamics of communities that are gaining influence; that act as bridging hubs; or isolated communities that are apparently losing influence. This analysis reveals factors that may underpin a scientific community’s growth or decline.

Extracted Key Phrases

6 Figures and Tables

Cite this paper

@inproceedings{Belk2015TheRO, title={The Risks of Introspection: A Quantitative Analysis of Influence Between Scientific Communities}, author={V{\'a}clav Bel{\'a}k and Conor Hayes}, booktitle={FLAIRS Conference}, year={2015} }