Answer Extraction from Technical Texts

Abstract

For most companies and organizations, technical documents are highly valued knowledge sources because they combine the know-how and experience of specialists in a particular domain. To guarantee the optimal use of these documents in specific problem situations, people must be able to quickly find precise and highly reliable information. Answer extraction is a new technology that helps users find precise answers to their questions in technical documents. In this article, the authors present ExtrAns, a real-world answer extraction system designed for technical domains. ExtrAns uses robust natural language processing technology and a semantic representation for information's propositional content. Knowing the forms of a domain's terminology and understanding the relation between the terms is vital for answer extraction. By applying rewrite rules in a systematic way, ExtrAns gets a grip on technical terminology.