LCC-DCU C-C Question Answering Task at NTCIR-5
Bin Wang, Gareth J. F. Jones · Arrow@dit (Dublin Institute of Technology) · 2005
This paper describes the work for our participation in the NTCIR-5 Chinese to Chinese Question Answering task. Our strategy is based on the “Retrieval plus Extraction” approach. We first retrieve relevant documents, then retrieve short passages from the above documents, and finally extract named entity answers from the most relevant passages. For question type identification, we use simple heuristic rules which can cover most questions. The Lemur toolkit with the OKAPI model is used for document retrieval. Results of our task submission are given and some preliminary conclusions drawn.