FAQ finder: a case-based approach to knowledge navigation

In this paper, we will discuss a class of systems, called FAQ FINDER systems, that use a natural language question-based interface to distributed information sources, specifically files organized as question/answer pairs such as FAQ files. In using these systems, users enter question in natural language and the system attempts answer that question, using FAQ files as a resource. We combined two technologies in developing these systems: statistically based IR engines and more semantically based “language matchers”. The power of our approach rises out of two features: first, we are using knowledge sources that have already been designed to answer the commonly asked questions in a domain and are more highly organized than free text; second, these systems do not have to comprehend the queries they receive. They only have to identify the relevant files and then match against the segments of text that are used to organize the files themselves