Analyzing Developer Sentiment throughout Development Process

Abstract

Emotions plays a major role in productivity, quality and job satisfaction. In this thesis, I have analyzed sentiments of developer on GitHub based on the comments of different open source projects and analyze their relationship with different factors such as day of week in which comment was made, time duration from which project was created and used programming language. I analyzed sentiment of those developers using comments written on issues, pull requests and commits. My results show that projects developed on Monday tend to a more negative emotion. Additionally, comments written in issues tend to have higher negative polarity in their emotional content and projects developed in java and python have more positive comments as compared to C and C++. There are many Sentiment Analysis Techniques available. I have also discussed about the Sentiment Analysis algorithms which can be used to improve accuracy and generate more accurate results. This thesis provides an overall approach to perform sentiment analysis and talks about the conclusions which can drawn from this thesis.