Abstract-- The performance of a retrieval system can be enhanced by integrating conventional Information Retrieval methods with techniques adapted from Artificial Intelligence. A prototype of such an Intelligent Information Retrieval System has been built at the University of Mississippi, using the determinants of medication subject domain as an application area. This paper explores how a natural language interface can increase recall and precision by eliminating the need for a formal query language and the use of trained intermediaries to formulate a request.