Abstract
The amount of time and money required to finish a software project and distribute the final product increases when there are bugs in the programme. Software procedures like defect monitoring and repair may be both costly and time-consuming to complete. Because it is difficult to locate and correct every defect in a product, it is essential that the negative effect of those defects be minimised in order to provide a result that is of better overall quality. The process of identifying troublesome sections of software code is known as software defect prediction. This paper presents an optimized machine learning-enabled model for software fault prediction to improve software quality. PC1 data set is fed as input data in this model. Important features are selected by ant colony optimization (ACO) technique. Selected features are fed as input to support vector machine. Training and testing of SVM is performed by PC1 data set. Performance of ACO SVM Ant Colony Optimization Support Vector Machine is compared with SVM, Naive Bayes classifier and K-Nearest Neighbour classifier. The performance of ACO-based SVM is better for software fault classification and prediction.