Metoda za analizo vpliva strukturnih sprememb izvorne kode na kakovost programske opreme : doktorska disertacija

slovenian

doktorska dela

2013

Univerza v Mariboru, Fakulteta za elektrotehniko računalništvo in informatiko

Software quality is an important characteristic in software development process. In order to precisely evaluate software quality and identify poor source code parts, researchers usually strive to use enhance mechanisms for quality assessment. Nowadays, the trends of software quality assessment and analysis are focusing on a longer time period in a software development process. In other words, software evolution represents a basis for software quality evaluation. In the dissertation, we are focusing on a software qualitythat is assessed from the software evolution. We are using structural source code changes and the concept of code churn in order to evaluate the quality. We used different machine learning algorithms to predict files with bugs. In order to test the usefulness of structural source code changes as a software quality indicator, we compared successfulness of predicting bugs withstructural source code changes and modified lines of code. The results show that structural source code changes can be used for software quality evaluation.

07.08.2023