Hybrid Approach to Software Fault Prediction Using Particle Swarm Optimization and Fuzzy Time Series

ABSTRACT Accurate software fault prediction is a critical aspect of proactive maintenance and overall software reliability enhancement. This paper introduces a model that combines particle swarm optimization (PSO) and fuzzy time series (FTS) to predict the number of faults present in the software. FTS, widely recognized for its forecasting capabilities, undergoes performance optimization in the proposed model through PSO, specifically by optimizing the length of intervals. The performance of the proposed model is evaluated on three real‐world software fault datasets sourced from open‐source software projects, revealing its superiority over alternative models. The evaluation is substantiated by consistently lower values of root mean absolute error, mean absolute error, and median absolute error. This research significantly contributes to the advancement of software reliability forecasting techniques, providing practitioners with a powerful tool for ensuring software reliability and enabling proactive measures for continuous improvement.