Abstract
Two of the main corpora available for training discourse relation classifiers are the RST Discourse Treebank (RST-DT) and the Penn Discourse Treebank (PDTB), which are both based on the Wall Street Journal corpus. Most recent work using discourse relation classifiers have employed fully-supervised methods on these corpora. However, certain discourse relations have little labeled data, causing low classification performance for their associated classes. In this paper, we attempt to tackle this problem by employing a semi-supervised method for discourse relation classification. The proposed method is based on the analysis of feature cooccurrences in unlabeled data. This information is then used as a basis to extend the feature vectors during training. The proposed method is evaluated on both RST-DT and PDTB, where it significantly outperformed baseline classifiers. We believe that the proposed method is a first step towards improving classification performance, particularly for discourse relations lacking annotated data.
| Original language | English |
|---|---|
| Title of host publication | Proceedings of the SIGDIAL 2010 Conference: 11th Annual Meeting of the Special Interest Group onDiscourse and Dialogue |
| Pages | 55-58 |
| Number of pages | 4 |
| Publication status | Published - 2010 |
| Externally published | Yes |
| Event | 11th Annual Meeting of the Special Interest Group on Discourse and Dialogue, SIGDIAL 2010 - Tokyo Duration: 2010 Sept 24 → 2010 Sept 25 |
Other
| Other | 11th Annual Meeting of the Special Interest Group on Discourse and Dialogue, SIGDIAL 2010 |
|---|---|
| City | Tokyo |
| Period | 10/9/24 → 10/9/25 |
ASJC Scopus subject areas
- Computer Graphics and Computer-Aided Design
- Computer Vision and Pattern Recognition
- Human-Computer Interaction
- Modelling and Simulation