A Two Phase Method to Predict the Software Faults using Self-organizing Map and Support-Vector Machine Algorithms

Software testing is one of the important activities to guarantee software reliability. But, software testing is a costly and time-consuming task in the process of software development. Predicting faulty modules of software before testing phase leads to reduce the cost of testing. In this paper, a two phase machine learning based method has been proposed to classify the software modules into faulty or non-faulty classes. In the proposed method a combination of self-organizing map (SOM) and support-vector machine (SVM) algorithms are used to create software module classifier. Regarding the results of experiments, the proposed method outperforms the previous machine-learning methods in terms of accuracy and prediction-error.

Keywords:

Software Fault Prediction , Self-organizing map , Support-vector machine , Accuracy , Prediction-error