Toward Fast and Accurate Neural Discourse Segmentation
This work addresses the problem of slow and impractical discourse segmenters for natural language processing researchers, though it is incremental as it builds on existing neural frameworks.
The paper tackles discourse segmentation by proposing an end-to-end neural segmenter using BiLSTM-CRF, which achieves new state-of-the-art performance on the RST-DT corpus and is significantly faster than previous methods.
Discourse segmentation, which segments texts into Elementary Discourse Units, is a fundamental step in discourse analysis. Previous discourse segmenters rely on complicated hand-crafted features and are not practical in actual use. In this paper, we propose an end-to-end neural segmenter based on BiLSTM-CRF framework. To improve its accuracy, we address the problem of data insufficiency by transferring a word representation model that is trained on a large corpus. We also propose a restricted self-attention mechanism in order to capture useful information within a neighborhood. Experiments on the RST-DT corpus show that our model is significantly faster than previous methods, while achieving new state-of-the-art performance.