Extracting Protein Terminologies in Literatures

Recently, key terminologies in literatures play an important role in analyzing and predicting research trends. Extracting those terminologies therefore used in the papers of researchers´ has become the most major issue in a variety of fields. To extract those terminologies, dictionary-based approach that contains terminologies has been applied. Wikipedia also can be considered as a dictionary since Wikipedia has abundant terminologies and power of the collective intelligence. It means that the terminologies are continuously modified and extended every day. Thus it could be an answer set to compare with the terminologies in literatures. However, it hardly extracts terminologies that are newly defined and coined by researchers. In order to solve this issue, we propose a method to derive a set of terminology candidates by comparing terminologies in literatures and Wikipedia. The candidate set extracted from the method showed an accuracy of about 64.33%, which is a good result as an initial study.