An Industrial Research Program in Software Fault Prediction

Abstract

It is often wondered why so much of the software engineering research that appears in the literature is not adopted by practitioners. After all, there are lots of exciting new ideas that could potentially improve both the quality and efficiency of software production. So why is this research ignored? Is it just a matter of ignorance or shortsightedness on the part of practitioners? In this paper we discuss our experience doing software fault prediction research in an industrial research lab, and describe the timeline, empirical studies and their results, and contrast them with a typical academic research program.