Horizontal traceability for JIT requirements

Agile projects typically employ just-in-time requirements engineering and record their requirements (socalled feature requests) in an issue tracker. In open source projects we observed large networks of feature requests that are linked to each other. Both when trying to understand the current state of the system and to understand how a new feature request should be implemented, it is important to know and understand all these (tightly) related feature requests. However, we still lack tool support to visualize and navigate these networks of feature requests. A first step in this direction is to see whether we can identify additional links that are not made explicit in the feature requests, by measuring the text-based similarity with a Vector Space Model (VSM) using Term Frequency - Inverse Document Frequency (TF-IDF) as a weighting factor. We show that a high text-based similarity score is a good indication for related feature requests. This means that with a TF-IDF VSM we can establish horizontal traceability links, thereby providing new insights for users or developers exploring the feature request space. The input data sets, the compiled stopword list and the raw cosine similarity Excel files are available for download.