Identification of Software Bugs by Analyzing Natural Language-BasedRequirements Using Optimized Deep Learning Features

Abstract

Software project outcomes heavily depend on natural language requirements, often causing diverse interpretationsand issues like ambiguities and incomplete or faulty requirements. Researchers are exploring machine learningto predict software bugs, but a more precise and general approach is needed. Accurate bug prediction is crucialfor software evolution and user training, prompting an investigation into deep and ensemble learning methods.However, these studies are not generalized and efficient when extended to other datasets. Therefore, this paperproposed a hybrid approach combining multiple techniques to explore their effectiveness on bug identificationproblems. The methods involved feature selection, which is used to reduce the dimensionality and redundancyof features and select only the relevant ones; transfer learning is used to train and test the model on differentdatasets to analyze how much of the learning is passed to other datasets, and ensemble method is utilized toexplore the increase in performance upon combining multiple classifiers in a model. Four National Aeronauticsand Space Administration (NASA) and four Promise datasets are used in the study, showing an increase in themodel's performance by providing better Area Under the Receiver Operating Characteristic Curve (AUC-ROC)values when different classifiers were combined. It reveals that using an amalgam of techniques such as those usedin this study, feature selection, transfer learning, and ensemble methods prove helpful in optimizing the softwarebug prediction models and providing high-performing, useful end mode