Community Formation and Detection on OSS Collaboration Networks

• Activity Number: 210 - SLDS CSpeed 3
• Type: Contributed
• Date/Time: Tuesday, August 10, 2021 : 1:30 PM to 3:20 PM
• Sponsor: Section on Statistical Learning and Data Science
• Author(s): Behnaz Moradi-Jamei* and Brandon L. Kramer and J. Bayoán Santiago Calderón and Gizem Korkmaz
• Companies: University of Virginia and University of Virginia and University of Virginia and University of Virginia
• Keywords: Community Detection; Community formation; Open Source Software; Networks

Abstract

This work studies community formation in OSS collaboration networks. We use a dataset of 3.26 million GitHub users and their repositories. While most current work examines the emergence of small-scale OSS projects, our work draws on a large-scale network of contributors. Moreover, OSS collaborations are characterized by small groups of users that collaborate closely, leading to the presence of short cycles of collaboration. To better understand how communities are shaped by the cyclic structure of the network rather than just the number of ties, we introduce a novel method for detecting communities that incorporates the aforementioned property as well as the strengths of the pairwise collaborations. Furthermore, we study factors that affect community formation, including user attributes such as the programming language of their choice, obtained from repositories languages assigned to users users’ countries. To do this, we propose and compare four assignment rules. This paper offers many insights for both open-source software experts and network scholars interested in studying open-source team formation.