PROJECT TITLE :

A Learning-to-Rank Approach to Software Defect Prediction

ABSTRACT:

Software defect prediction will help to allocate testing resources efficiently through ranking software modules according to their defects. Existing software defect prediction models that are optimized to predict explicitly the quantity of defects in a software module would possibly fail to grant an correct order as a result of it is terribly tough to predict the exact variety of defects in a very software module due to noisy information. This paper introduces a learning-to-rank approach to construct software defect prediction models by directly optimizing the ranking performance. In this paper, we tend to build on our previous work, and further study whether the thought of directly optimizing the model performance measure can benefit software defect prediction model construction. The work includes two aspects: one may be a novel application of the educational-to-rank approach to real-world data sets for software defect prediction, and the other could be a comprehensive evaluation and comparison of the educational-to-rank technique against other algorithms that have been used for predicting the order of software modules in line with the anticipated variety of defects. Our empirical studies demonstrate the effectiveness of directly optimizing the model performance live for the learning-to-rank approach to construct defect prediction models for the ranking task.