Matching Questions and Answers in Dialogues from Online Forums
This work addresses a key step in dialogue analysis and system training for online forums, but it is incremental as it builds on existing attention mechanisms and focuses on a specific dataset.
The paper tackled the problem of matching questions and answers in online forum dialogues by proposing a model that uses mutual attention to incorporate distance and dialogue history, and demonstrated that it outperforms state-of-the-art baselines, especially for long-distance pairs, on a newly created dataset of 1,000 labeled dialogues.
Matching question-answer relations between two turns in conversations is not only the first step in analyzing dialogue structures, but also valuable for training dialogue systems. This paper presents a QA matching model considering both distance information and dialogue history by two simultaneous attention mechanisms called mutual attention. Given scores computed by the trained model between each non-question turn with its candidate questions, a greedy matching strategy is used for final predictions. Because existing dialogue datasets such as the Ubuntu dataset are not suitable for the QA matching task, we further create a dataset with 1,000 labeled dialogues and demonstrate that our proposed model outperforms the state-of-the-art and other strong baselines, particularly for matching long-distance QA pairs.