Abstract

Modern information retrieval technology is playing an increasingly important role in our daily life. Keyword based search engine system and question answering system are the two main technologies of text information retrieval on the Internet. Current search engine systems do not give enough contexts for their search, and users have to go through the documents to find relevant information. Due to their complexity, question answering systems typically have slow response time. The Virtual Question Answering System (VQAS) presented in this thesis aims to provide a method to retrieve full-text information from a digital library by targeting the gap between a true question answering system and current search engines. It retrieves the paragraphs where the possible response corresponding to the user's query appear, not just a few words before and after the search terms. In most cases, the user would be able to find the answer directly in the paragraphs of the query result.