DOI: https://doi.org/10.15368/theses.2016.28
Available at: https://digitalcommons.calpoly.edu/theses/1568
Date of Award
4-2016
Degree Name
MS in Computer Science
Department/Program
Computer Science
Advisor
Davide Falessi
Abstract
Organizations typically use issue tracking systems (ITS) such as Jira to plan software releases and assign requirements to developers. Organizations typically also use source control management (SCM) repositories such as Git to track historical changes to a code-base. These ITS and SCM repositories contain valuable data that remains largely untapped. As developers churn through an organization, it becomes expensive for developers to spend time determining which software artifact must be modified to implement a requirement. In this work we created, developed, tested and evaluated a tool called Class Change Predictor, otherwise known as CCP, for predicting which class will implement a requirement. Understanding which class will implement a requirement supports several software engineering tasks such as refactoring and assigning requirements to developers.
CCP is a data-mining tool operating on top of ITS and SCM repositories which gathers a unique combination of metrics. CCP leverages requirement text to compare current requirements to past requirements and requirements to source code files. CCP performs static analysis on the code-base of each major release of the software artifact. We evaluated CCP on different open source datasets (and the Digital Democracy dataset) by using several machine learning classifiers and pre-processing procedures. Our results show that we can achieve high precision on three out of four datasets. We conclude that accurate class change prediction is feasible, and we propose numerous solutions to increase future accuracy.