The University of Amsterdam at TAC 2008 Question Answering Track
暂无分享,去创建一个
We describe the participation of the University of Amsterdam's ILPS group in the Question Answering track of the TAC 2008. We used a simple system based on lexicon-based identification of opinionated sentences and the answer extraction module of our factual question answering system. The results indicate that filtering out sentences that are unlikely to contain opinions does improve the end-to-end performance of the system.
[1] Janyce Wiebe,et al. Recognizing Contextual Polarity in Phrase-Level Sentiment Analysis , 2005, HLT.
[2] Katja Hofmann,et al. The University of Amsterdam's Question Answering System at QA@CLEF 2007 , 2008, CLEF.
[3] M. de Rijke,et al. The University of Amsterdam at TREC 2008: Blog, Enterprise, and Relevance Feedback , 2008 .