Test Case Prioritization Using Swarm Intelligence Algorithm to Improve Fault Detection and Time for Web Application

Abstract

Prioritizing test cases based on several parameters where important ones are executed first is known as test case prioritization (TCP). Code coverage, functionality, and features are all possible factors of TCP for detecting bugs in software as early as possible. This research was carried out to test and compare the effectiveness Swarm Intelligence algorithms, where Artificial Bee Colony (ABC) and Ant Colony Optimization (ACO) algorithms were implemented to find the fault detected and execution time as these are the curial aspects in software testing to ensure good quality products are produced within the timeline. As web applications are commonly used by a board population, this research was carried out on an Online Shopping application represented as Case Study One and Education Administrative application known as Case Study Two. In recent years, TCP has been implemented widely, but none has implemented on web application which was conducted to fill the gaps and produce a new contribution in this area. The outcome was compared using Average Percentage Fault Detected (APFD) and execution time. For Case Study One, the APFD value was 0.80 and 0.71 while the execution time was 8.64 seconds and 0.69 seconds respectively for ABC and ACO. For Case Study Two, the APFD values were 0.81 and 0.64 while the execution time was 8.83 seconds and 1.22 seconds for ABC and ACO. It was seen that both algorithms performed well in their respective ways. ABC had shown to give a higher value for APFD while ACO had converged faster for execution time.