Abstract

Machine learning and statistical techniques are used in software fault prediction to predict the presence or the absence of faults in the given software modules. In order to make the predictions, a software fault prediction learns upon the software fault data having the information about the software system (software metrics) augmented with the fault value. An implicit requirement to perform effective software fault prediction is the availability of reasonable quality fault data. However, obtaining quality software fault data is difficult as in general software development companies are not keen to share their software development information or they are not having any software repository in first place.

Full Text
Paper version not known

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call