Extraction Active Ingredient Names from Medical Package Inserts

Abstract

The development of drug database is needed to ensure medical safety. However, the actual database does not have a suitable data structure which is applied to medical information system since the drug information is not well organized. As the first step of development of well organized database based on medical package inserts which have information such as drug names, efficacy and dosage, in this study, we propose the method to extract active ingredient names from SGML data of the package inserts using extended pattern matching.