Software defect prediction (SDP) has been a very popular research topic in software engineering for the past two decades. Its goal is to find bugs and errors in the early stages of the software development life cycle (SDLC). Traditional machine learning algorithms cannot make the software easier to test, maintain, or use. To overcome these challenges, we have proposed a hybrid AOA-MLP model. The proposed model uses feature selection and defect prediction, which are both important for improving predictive performance. The AOA is used to choose features while ML is used to find defects. This model overcomes critical issues in existing SDP techniques, including excessive time complexity and the curse of dimensionality, by efficiently selecting meaningful features out of a large set of potential predictors. This research ensures betterment to the field of software quality assurance by use of machine learning techniques, particularly feature reduction strategies, to enhance the accuracy and relevance of software defect prediction.