Abstract
The most expressive attribute for defining successful software is its quality, which can only be attained when the chances of a defect occurrence are doubtful. Software defect Prediction is a cycle to develop a model that software practitioners can use to detect faulty areas before the testing phase. Prediction of such classes before its testing phase will enable the development team to allocate resources more efficiently. It reduces the testing effort and can yield higher quality software with reduced cost. Prediction of such defects can be further elevated when machine learning techniques are introduced to such a Software defect Prediction (SDP) model. In this paper, we conducted a systematic literature review of the studies that are available on such practices of predicting defects in software before the actual testing is started with a blend of machine learning techniques that provide an edge to software defect prediction model is carried out with the latest Research present in this area providing the most upgrade statistics on Software defect prediction using machine learning.