Abstract

Software operates various essential systems and devices in today's world. Many companies build systems with varying sizes and functions in order to offer higher-quality software more quickly. The goal of this diversity in software development is to effectively meet the goals of the client while adjusting to the many demands of the current technological environment. Early software bug prediction improves software success overall by supporting quality, dependability, efficiency, and cost reduction—all important aspects of software development and maintenance. Through a careful analysis of the collection of existing research, the goal is to investigate current trends in early software bug prediction. The review concludes by highlighting broad range of approaches for machine learning including Neural Networks, Random Forest, Logistic Regression, and Naïve Bayes with software metrics like source code metrics etc. The performance and reliability of the model are assessed through the use of metrics like accuracy, recall, precision and F1-Measure.