Natural language processing is the subdiscipline within Computer Science that focuses on handling human language, like English, as a means of communication rather than as a string of characters. Language, as humans use it, has structure (grammar and syntax) and content (vocabulary and underlying knowledge) so processing natural language requires delineation of structural and content elements as a starting point. Writing parsers, defining grammars, building lexicons and knowledge bases, and putting all the pieces together into software systems -- that�s natural language processing! But NLP also involves statistical analysis of language patterns, discourse analysis, text generation, and speech processing. Understanding language also requires an ability to reason and understand common cultural references -- capabilities we humans take for granted.

Background

Bio: Mary Dee Harris received her Ph.D. degree in English literature and Computer Science from the University of Texas at Austin. Her career includes 15 years teaching Computer Science, most recently at Loyola University in New Orleans, and over ten years in the computer industry. Currently Dr. Harris runs her own consulting business, Language Technology, Inc., specializing in applying natural language processing and knowledge systems solutions to problems of handling natural language text in software systems. Her clients have included the National Science Foundation, the U.S. Army, and Educational Testing Service (ETS). Dr. Harris�s work has focused on many areas in Natural Language Processing, especially the design of natural language systems and the development of lexical knowledge bases. Over the last five years, she has advised ETS on automatic scoring of natural language responses to test questions, designed to replace multiple choice test questions generally used now.

CREST home page	News	Research	Seminar Series	Publications	Feedback