Towards emotional awareness in software development teams

Abstract

Emotions play an important role in determining work results and how team members collaborate within a project. When working in large, distributed teams, members can lose awareness of the emotional state of the project. We propose an approach to improve emotional awareness in software development teams by means of quantitative emotion summaries. Our approach automatically extracts and summarizes emotions expressed in collaboration artifacts by combining probabilistic topic modeling with lexical sentiment analysis techniques. We applied the approach to 1000 collaboration artifacts produced by three development teams in a three month period. Interviews with the teams' project leaders suggest that the proposed emotion summaries have a good correlation with the emotional state of the project, and could be useful for improving emotional awareness. However, the interviews also indicate that the current state of the summaries is not detailed enough and further improvements are needed.