DMCA
Syntactic and semantic structure for opinion expression detection (2010)
Cached
Download Links
| Venue: | In Proceedings of the 14th Conference on Computational Natural Language Learning |
| Citations: | 21 - 4 self |
BibTeX
@INPROCEEDINGS{Johansson10syntacticand,
author = {Richard Johansson and Ro Moschitti},
title = {Syntactic and semantic structure for opinion expression detection},
booktitle = {In Proceedings of the 14th Conference on Computational Natural Language Learning},
year = {2010}
}
Share
 |
 |
 |
 |
||
OpenURL
Abstract
We demonstrate that relational features derived from dependency-syntactic and semantic role structures are useful for the task of detecting opinionated expressions in natural-language text, significantly improving over conventional models based on sequence labeling with local features. These features allow us to model the way opinionated expressions interact in a sentence over arbitrary distances. While the relational features make the prediction task more computationally expensive, we show that it can be tackled effectively by using a reranker. We evaluate a number of machine learning approaches for the reranker, and the best model results in a 10-point absolute improvement in soft recall on the MPQA corpus, while decreasing precision only slightly. 1
Keyphrases
semantic structure opinion expression detection relational feature mpqa corpus machine learning approach natural-language text local feature semantic role structure arbitrary distance model result 10-point absolute improvement conventional model soft recall prediction task opinionated expression way opinionated expression
