The primary functionality provided by the MCE is the extraction of concepts from various types of multi-media information source units. The framework of the MCE is designed to support both text and image source units. Currently, a prototype of the MCE has been implemented with a foundation for processing text-based information sources. The MCE supports concept extraction from documents of various formats including, for example, SGML. It employs advanced natural language parsing techniques to identify noun phrases and then extracts said noun phrases and stores them as concepts.