Date of Award
Daniel Felix Ritchie School of Engineering and Computer Science, Computer Science
Anneliese Amschler Andrews
Change request, Effort, Estimation, Evolution, Legacy system, Prediction
Prediction of software defects has been the focus of many researchers in empirical software engineering and software maintenance because of its significance in providing quality estimates from the project management perspective for an evolving legacy system. Software Reliability Growth Models (SRGM) have been used to predict future defects in a software release. Modern software engineering databases contain Change Requests (CR), which include both defects and other maintenance requests. Our goal is to use defect prediction methods to help predict CRs in an evolving legacy system.
Limited research has been done in defect prediction using curve-fitting methods evolving software systems, with one or more change-points. Curve-fitting approaches have been successfully used to select a fitted reliability model among candidate models for defect prediction. This work demonstrates the use of curve-fitting defect prediction methods to predict CRs. It focuses on providing a curve-fit solution that deals with evolutionary software changes but yet considers long-term prediction of data in the full release. We compare three curve-fit solutions in terms of their ability to predict CRs. Our data show that the Time Transformation approach (TT) provides more accurate CR predictions and fewer under-predicted Change Requests than the other curve-fitting methods.
In addition to CR prediction, we investigated the possibility of estimating effort as well. We found Lines of Code (added, deleted, modified, and auto-generated) associated with CRs do not necessarily predict the actual effort spent on CR resolution.
Copyright is held by the author. User is responsible for all copyright compliance.
Lamees Abdullah Alhazzaa
Received from ProQuest
Alhazzaa, Lamees Abdullah, "Change Request Prediction and Effort Estimation in an Evolving Software System" (2021). Electronic Theses and Dissertations. 1888.