Developing a lexicon for automatic knowledge acquisition

The major of the bottleneck in expert system development is knowledge acquisition. This paper focuses on the development of a comprehensive lexicon to be used in an automated knowledge acquisition system. Previous work demonstrated the feasibility of automatically acquiring expert knowledge from natural language analysis of articles published in medical journals. Domain specific knowledge was extracted from English text and then converted into a rule-based knowledge representation for a medical expert system. Previous work used a domain specific sublanguage lexicon that was focused on the specific domain area. The current work is focused on developing a more general purpose lexicon that can be used for a deeper understanding of the knowledge contained in English text