Towards Semi-automatic Ontology Building Supported by Large-scale Knowledge Acquisition
Yi‐Min Wang, Johanna, Peter N. Haase · 2006
Knowledge acquisition is usually the first step in building ontologies. On the one hand, knowledge is typically im-plicitly contained in large collections of unstructured docu-ments. Therefore it is extremely troublesome to manually identify relevant concepts. On the other hand, users are of-ten not fully satisfied with the results of automated state-of-the-art ontology learning techniques. In this paper we present a technique for large-scale Knowledge Acquisition supported Semi-automated Ontology building (KASO) and a corresponding software system. By applying KASO and using this software, users are able to bootstrap the process of building high quality ontologies by automatically acquir-ing concepts from large-scale document collections and to make use of traditional knowledge acquisition approaches to refine and organize the machine-generated concepts. Evalua-tion studies and user experiences indicate the applicability of KASO in bootstrapping ontology construction.