ABSTRACT
Software quality is the key aspect of every software organization. Multiple frameworks and algorithms are essential to ensure quality. However, multiple software failures occur uninvited. There are multiple aspects that skew a software’s efficiency. Now the software quality analysis framework mostly focuses on design flaws and test plans done during development. To overcome this problem of software failure, this research proposes a prediction for software efficiency analysis in software engineering using enhanced feed-forward neural network machine learning classification with CatBoost. This research also evaluates the parameters of efficiency of each software component before implementation. This proposed work also analyses the basic aspects that need to be ensured before the design phase of any software.
