A Comparison of Rule-Based and Machine Learning Methods for Medical Information Extraction
Osamu Imaichi, Toshihiko Yanase, Yoshiki Niwa · 2013
This year's MedNLP (Morita and Kano, et al., 2013) has two tasks: de-identification and complaint and diagnosis. We tested both ma-chine learning based methods and an ad-hoc rule-based method for the two tasks. For the de-identification task, the rule-based method achieved slightly higher results, while for the complaint and diagnosis task, the machine learning based method had much higher re-calls and overall scores. These results suggest that these methods should be applied selective-ly depending on the nature of the information to be extracted, that is to say, whether it can be easily patternized or not. 1