Biomedical term extraction using fuzzy association

Automatic term extraction from a biomedical text is a well-known problem in the area of natural language processing. It is carried out by employing four kinds of measures: linguistic and rule-based, dictionary-based, statistical, and machine learning. Automatic term extraction indicates whether or not two or more words come together in the text more often than by chance to form a biomedical term that is automatically extracted using an automated system. A fuzzy set-theoretic approach is presented in this article and compares with the existing statistical measures. The experimental result shows that the fuzzy measure offers better precision than the popular statistical measures for extracting biomedical terms, especially when we have compared more than 60% ranked list of extracted terms.