A Trainable Multi-factored QA System

This paper reports on the construction and testing of a new Question Answering (QA) system, implemented as an workflow which builds on several web services developed at the Research Institute for Artificial Intelligence (RACAI).The evaluation of the system has been independently done by the organizers of the RomanianRomanian task of the ResPubliQA 2009 exercise. We further describe a principled way of combining different relevance factors used for computing a general score, based on which are returned the paragraphs which are most likely to contain the correct answers to the asked questions. The system was trained on a specific parallel corpus, but its functionality is independent on the linguistic register of the training data. The trained QA system which participated in the ResPubliQA shared task is available as a new web service on the RACAI's web-services page. ACM Categories and Subject Descriptors: H.3.1 Content Analysis and Indexing

• Type: Conference
• Year: 2009
• Where: CLEF
• Authors: Radu Ion, Dan Stefanescu, Alexandru Ceausu, Dan Tufis, Elena Irimia, Verginica Barbu Mititelu
• Keywords: CLEF 2009, Content Analysis, Information Search, Information Technology, Web Service