A Practitioner Approach of Deep Learning Based Software Defect Predictor

Abstract

In Software Development Life Cycle (SDLC), the coding plays the very crucial phase so far as quality of software is concerned. The quality of software highly depends  on the quality of coding done by the software developer. Minor defects in software may results in huge loss to software development firm. To test phase of the software development life cycle is very much required, it is a mechanism of  quality control system in SDLC. Early detection of defect in software, mostly during development saves the time of testing and increase the development efficiently. There are lots of ML Model developed by researchers for said purpose. Natural Language Processing (NLP) based Software Defect Predictor outperformed the Traditional ML based techniques. Also Deep Learning based feature extractor has outperformed the hand crafted Software metric.