A large scale Linux-Kernel based benchmark for feature location research

Abstract

Many software maintenance tasks require locating code units that implement a certain feature (termed as feature location). Feature location has been an active research area for more than two decades. However, there is lack of publicly available, large scale benchmarks for e valuating and comparing feature location approaches. In this paper, we present a LinuxKernel based benchmark for feature location research. This benchmark is large scale and extensible. By providing rich feature and program information and accurate ground-truth links between features and code units, it supports the e valuation of a wide range of feature location approaches. It allows researchers to gain deeper insights into existing approaches and how they can be improved. It also enables communication and collaboration among different researchers. (video: http://www.youtube.com/watch?v=3D_HihwRNeK3I).