Train One Get One Free: Partially Supervised Neural Network for Bug Report Duplicate Detection and Clustering
This work addresses the engineering effort in tracking user-reported bugs by automating duplicate detection and clustering, though it is incremental as it builds on existing neural methods for these tasks.
The paper tackles the problem of duplicate bug report detection and clustering by proposing a neural architecture that jointly performs both tasks using only duplicate classification supervision. The results show that the model outperforms state-of-the-art methods in duplicate classification on real-world datasets and learns meaningful latent clusters without extra supervision.
Tracking user reported bugs requires considerable engineering effort in going through many repetitive reports and assigning them to the correct teams. This paper proposes a neural architecture that can jointly (1) detect if two bug reports are duplicates, and (2) aggregate them into latent topics. Leveraging the assumption that learning the topic of a bug is a sub-task for detecting duplicates, we design a loss function that can jointly perform both tasks but needs supervision for only duplicate classification, achieving topic clustering in an unsupervised fashion. We use a two-step attention module that uses self-attention for topic clustering and conditional attention for duplicate detection. We study the characteristics of two types of real world datasets that have been marked for duplicate bugs by engineers and by non-technical annotators. The results demonstrate that our model not only can outperform state-of-the-art methods for duplicate classification on both cases, but can also learn meaningful latent clusters without additional supervision.