Mining intentions to improve bug report summarization

Abstract

In recent years, various automatic summarization techniques have been proposed to extract important information from bug reports. However, existing techniques mainly focus on common text features and ignore human intentions implied in bug reports. In fact, each bug report generally contains multiple intentions which are distributed in different sentences. Bug report readers are usually more interested in content that contains intentions of certain categories (e.g. fix solution, bug description). Based on the above observation, we introduce an intention taxonomy and implement the intention classification algorithm in this paper. Furthermore, we propose a new Intention-based Bug Report Summarization approach, namely IBRS, which leverages intention taxonomy to enhance bug report summarization. We evaluate our approach on Intention-BRC corpus and the experimental result shows that IBRS outperforms the state-of-The-Art approaches in terms of precision, recall, F-score, and pyramid precision.