Using information extraction and natural language generation to answer e-mail (Q2734655)

This paper discusses the use of information extraction and natural language generation in the design of an automated e-mail answering system. We analyse short free-form texts and generate a customised and linguistically motivated answer to frequently asked questions. We describe the approach and the design of a system currently being developed to answer e-mail in French regarding printer-related questions addressed to the technical support staff of our computer science department. A domain-dependent approach was preferred because precision is crucial in e-mail answering, as a wrong answer sent to a client can have negative consequences in terms of customer satisfaction.