Defect Classification Using Naive Bayes Classification

摘要

In software development life cycle (SDLC) identifying defects and classifying them as Major, Moderate and Minor is important to find the impact on effort, schedule and cost. Classification is a data mining task to assign a data item to a predefined set of classes. The performance of classifier depends upon the type of defects identified in various stages by the class label of severity in the training data. In this research work, Naive Bayes classification is used to predict the class label of "severity" tuple. The data tuples are described by the attributes of Defect, Phase Detected, Phase Attributed, Impact and Weight.