Systematization of Medical Terminology Knowledge Based on Examples

Abstract

　In medical information system, natural language processing interface will be useful if it realize. However, “Terminology Problem” arise. “Terminology Problem” have many aspects, such as standardization and lexicon harmonization. In order to find a solution to the problems, we focused on the systematization of terminological knowledge which is based on actual medical text “The Merck Manual.” In order to find out what kinds of terms are distinctively used in the medical knowledge architecture that The Merck Manual has, an automatic indexing method “TF/IDF weighting.” It was applied to perform statistically approximate calculation with the big medical terminology glossary which made by ours and thus, terms which depend of fields were extracted to develop a systematized lexicon.