APPARATUS AND METHOD FOR SEARCHING INFORMATION BASED ON WIKIPEDIA'S CONTENTS

摘要

The present invention relates to an information searching apparatus based on the Wikipedia content and an information searching method. An embodiment of the present invention provides the information searching apparatus based on the Wikipedia content comprises: a document conversion unit extracting a body document, a section title document, an info-box document, a category document and a definition statement document from the original text of the Wikipedia to generate one or more Wikipedia documents for response to a query; a document indexing unit analyzing the Wikipedia documents for response to the query to extract index words of a word class unit from the Wikipedia documents for response to the query and generating the index of the Wikipedia documents for response to the query; a query analyzing unit receiving a natural language query, analyzing a query type, an answer type and a query focus and extracting document search keywords; a document searching unit performing a document search from the Wikipedia documents for response to the query using the document search keywords and generating a document search result from the index of the Wikipedia documents for response to the query; an answer extracting unit extracting a first answer from the document search result using information on the query type, the answer type and the query focus; and an answer integration unit integrating the first answer and ranking the first answer to generate a second answer.