Robust Question Answering for Speech Transcripts Using Minimal Syntactic Analysis

Abstract. This paper describes the participation of the Technical University of Catalonia in the CLEF 2007 Question Answering on Speech Transcripts track. For the processing of manual transcripts we have deployed a robust factual Question Answering that uses minimal syntactic information. For the handling of automatic transcripts we combine the QA system with a novel Passage Retrieval and Answer Extraction engine, which is based on a sequence alignment algorithm that searches for “sounds like” sequences in the document collection. We have also enriched the NERC with phonetic features to facilitate the recognition of named entities even when they are incorrectly transcribed. Key words: Question Answering, Spoken Document Retrieval, Phonetic Distance

• Type: Conference
• Year: 2007
• Where: CLEF
• Authors: Pere Comas, Jordi Turmo, Mihai Surdeanu
• Keywords: CLEF 2007, Clef 2007 Question, Information Retrieval, Question Answering, Robust Factual Question