Comparison of Feature Selection via Semi supervised denoising autoencoder and traditional approaches For Software Fault-prone Classification

Abstract

Software quality is the capability of a software process to produce software product satisfying the end user. The quality of process or product entities is described through a set of attributes that may be internal or external. For the product entity, especially, the source code, different internal attributes are defined to evaluate its quality like complexity and cohesion. Concerning external attributes related to the product environment like reliability, their assessment is more difficult. Thus, they are usually predicted by the development of prediction models based on software metrics as independent variables and other measurable attributes as dependent variables. For instance, reliability like other external attributes is generally measured and predicted based on other quality attributes like defect density, defect count and fault-proneness. The success of machine learning (ML) and deep learning (DL) approaches for software defect and faulty modules classification as crucial attributes for software reliability improvement is remarkable. In recent years, there has been growing interest in exploring the use of deep learning autoencoders, a type of neural network architecture, for software defect prediction. Therefore, we aim in this paper to explore the semi-supervised denoising DL autoencoder in order to capture relevant features. Then, we evaluate its performance in comparison to traditional ML supervised SVM technique for fault-prone modules classification. The performed experiments based on a set of software metrics extracted from NASA projects achieve promising results in terms of accuracy and show that denoising DL autoencoder outperforms traditional SVM technique.