• Madhusmita Das
• Debasish Pradhan
• Suryakant Mahapatra

The process of software testing is crucial in the development of software. Usually, errors committed by developers are corrected during the latter phases of the software development procedure., resulting in a greater impact from the defects. To avoid this, it is essential to anticipate defects early on during the software development phase. This proactive approach allows for the efficient allocation of testing resources. The process of defect prediction entails categorizing software modules into those likely to have defects and those not likely to have defects. The main goal of this study is to reduce the negative effects caused by two major issues faced in defect prediction, namely, the unequal distribution of data and the extensive number of factors in defect datasets. This research paper involves assessing multiple software metrics using feature selection methods like PCA, along with several machine learning classifiers including Adaboost, MLP, NB, J48, and Random Forest. The objective is to classify software modules as either prone to defects or not prone to defects. The suggested model utilizes a blend of Adaboost and random forest classifiers, incorporating PCA for dimension reduction. The experimental analysis relies on the publicly available NASA dataset. The findings indicate that the combination of Adaboost and random forest algorithms, coupled with PCA for the MC1 dataset, produced the most favourable outcomes compared to other datasets. The defect prediction accuracy reached an impressive 98.52% in contrast to the algorithms employed in the study.

Software fault Prediction(SFP), AdaBoost, Random Forest, ADRF, Principal Component Analysis(PCA)

"A PCA BASED SOFTWARE FAULT PREDICTION MODEL USING ADRF", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.10, Issue 6, page no.j189-j199, June-2023, Available :http://www.jetir.org/papers/JETIR2306926.pdf

"A PCA BASED SOFTWARE FAULT PREDICTION MODEL USING ADRF", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.10, Issue 6, page no. ppj189-j199, June-2023, Available at : http://www.jetir.org/papers/JETIR2306926.pdf