Discovering Hidden Knowledge from Biomedical Literature

In this paper we investigate the potential of text mining for discovering implicit knowledge in biomedical literature. Based on Swanson's suggestion for hypotheses generation we tried to identify potential contributions to a better understanding of autism focusing on articles from database PubMed Central. First, we used them for ontology construction in order to obtain an improved insight into the domain structure. Next, we extracted a few rare terms that could potentially lead to new knowledge discovery for the explanation of the autism phenomena. We present a concrete example of such constructed knowledge about a substance calcineurin and its potential relations with other already published indications of autism. Povzetek: Prispevek opisuje uporabo metod rudarjenja besedil na medicinskih člankih s področja avtizma.