Toward Extractive Summarization of Online Forum Discussions via Hierarchical Attention Networks

Forum threads are lengthy and rich in content. Concise thread summaries will both newcomers seeking information and those who participate in the. Few studies, however, have examined the task of forum thread. In this work we make the first attempt to adapt the hierarchical networks for thread summarization. The model draws on the recent of neural attention mechanisms to build sentence and thread and use them for summarization. Our results indicate that the approach can outperform a range of competitive baselines. Further, a removal step is crucial for achieving outstanding results.