Development of Robust Traceability Benchmarks

Traceability benchmarks are essential for the evaluation of traceability recovery techniques. This includes the validation of an individual traceability technique itself and the objective comparison of the technique with other traceability techniques. However, it is generally acknowledged that it is a real challenge for researchers to obtain or build meaningful and robust benchmarks. This is because of the difficulty of obtaining or creating suitable benchmarks. In this paper, we describe an approach to enable researchers to establish affordable and robust benchmarks. We have designed rigorous manual identification and verification strategies to determine whether or not a link is correct. We have developed a formula to calculate the probability of errors in benchmarks. Analysis of error probability results shows that our approach can produce high quality benchmarks, and our strategies significantly reduce error probability in them.