A Passage Retrieval System for Multilingual Question Answering

In this paper we present a new Question Answering (QA) system based on redundancy and a new Passage Retrieval (PR) method oriented to QA. We suppose that in a large enough document collection we can find the answer to any question of several forms. Therefore, it is possible to find one or several sentences including the answers which contain part of the original questions. Next, we use Support Vector Machine (SVM) and Pattern Matching (PM) to identify the answer type and extract the answer respectively.

• Type: Conference
• Year: 2005
• Where: TSD
• Authors: José Manuel Gómez Soriano, Manuel Montes-y-Gómez, Emilio Sanchis Arnal, Paolo Rosso
• Keywords: Large Enough Document, Question Answering, Signal Processing, Support Vector Machine, TSD 2005