Detection Bad Code Smells By Using Deep Machine Learning Approaches

Code smells are an indication of deviation from design principles or implementation in the source code. Early detection of these code smells increases software quality. In contemporary times, diverse artificial neural network models are employed in software engineering to detect code smells without the need for source code access, but only large training sets. The aim of a study is to present a Recurrent Neural Network, Long Short Term Memory and Gated Recurrent Units models for detecting three types of code smells (Feature Envy, Long Method, and God Class) based on Java projects. Three code smell datasets were considered. Various performance metrics were employed for comparing the results. The experimental findings suggest that the proposed models demonstrate superior performance in identifying code smells, and augmenting the training data can improve the accuracy of predictions.