Overview of knowledge extraction techniques in five question-answering systems

Knowledge extraction is a complex process allowing the identification of previously unknown structures and potentially useful original information from big masses of data. Among its domains of application, we find the question answering systems. This article aims to provide an overview of five Q/A systems : IntelliServe, LaSIE, Quantum, SBLD and Watson. IntelliServe and Watson are commercial and the others are dedicated to the field of research. Each of these systems uses special techniques for the treatment of requests as well as for the extraction of the response.