Automatic labelling of documents based on ontology

Abstract

Recently cloud systems have been spread widely through internet so that we can get to huge amount of complex information (mainly in text) easily and quickly. However we can hardly catch up with the changes inside since we should go through them and see quickly what's going on. This is why automatic labelling is indispensable. In this work, assuming concept hierarchy, we extract suitable labels for documents by abstracting and ranking characteristic words.