Deep convolutional neural network model for bad code smells detection based on oversampling method

Code <span>smells refers to any symptoms or anomalies in the source code that shows violation of design principles implementation. Early detection bad smells improves software quality. Nowadays several artificial neural network (ANN) models have been used for different topics engineering: defect prediction, vulnerability detection, and clone detection. It is not necessary know data when using ANN but require large training sets. Data imbalance main challenge intelligence techniques detecting smells. To overcome these challenges, objective this study presents deep convolutional (D-CNN) model with synthetic minority over-sampling technique (SMOTE) detect based on a set Java projects. We considered four code-smell datasets which are God class, feature envy long method results were compared performance measures. Experimental show proposed oversampling can provide better prediction be further improved trained more datasets. Moreover, epochs hidden layers help increase accuracy model.</span>