Robin D. Burke, Kristian J. Hammond, and Edwin Cooper
This paper describes FAQ FINDER, a natural language question-answering system that uses files of frequently-asked questions as its knowledge base. Unlike AI question-answering systems that focus on the generation of new answers, FAQ FINDER retrieves existing ones found in frequently-asked question files. Unlike information retrieval approaches that rely on a purely lexical metric of similarity between query and document, FAQ FINDER uses a semantic knowledge base (WordNet) and natural language processing techniques to improve its ability to match question and answer. We describe an evaluation of the system’s performance against a corpus of user questions, and show that a combination of techniques from information retrieval and natural language processing works better than any single approach.