E4 - Introduction to Information Extraction Technology

Douglas Appelt & David Israel

Information Extraction (IE) Technology is directed at recovering specific, highly structured information from ordinary natural language texts such as newspaper articles, email messages, and other on-line sources of textual information. Recent research has led to advances in IE technology that make this collection of techniques ripe for practical application. In this tutorial, we will discuss

1. the nature of the technology, and what distinguishes it from related areas
2. the types of applications for which the technology is suitable
3. techniques for the evaluation of IE system performance
4. an analysis of a typical IE system and its components
5. an overview of both theoretical and practical techniques relevant to building information extraction systems
6. where to find public domain resources for building information extraction systems, and
7. limitations of the technology, and productive areas for future research.

Prerequisite knowledge:
We do not make any assumption about the background of tutorial attendees other than a general familiarity with computing, automata, grammars, and languages that would be typical of an undergraduate computer science major.

Douglas Appelt is a senior computer scientist in the Artificial Intelligence Center of SRI International. He received his Ph.D. in Computer Science in 1981 from Stanford University. Dr. Appelt has conducted research in natural-language generation, speech acts, and spoken language systems. He is affiliated with the Center for the Study of Language and Information at Stanford University, and was president of the Association for Computational Linguistics and has served on the editorial boards of Computational Linguistics and Computational Intelligence.

David Israel is a senior computer scientist in the Artificial Intelligence Center of SRI International. He received his Ph.D. in Philosophy from the University of California at Berkeley in 1974. He has served on the faculty of Tufts University, and as a research scientist at BBN. Dr. Israel has been active in research in natural language semantics, formal logic and knowledge representation, and has been involved in numerous projects at SRI involving the application of information extraction technology. He is on the editorial board of Computational Intelligence.