Designing Clustering Methods for Ontology Building: The Mo'K Workbench

TitleDesigning Clustering Methods for Ontology Building: The Mo'K Workbench
Publication TypeConference Paper
Year of Publication2000
AuthorsBisson, G, Nédellec, C, Cañamero, LD
EditorStaab, S, Maedche, A, Nédellec, C, Wiemer-Hastins, P
Name of ProceedingsProc. First Workshop on Ontology Learning. Workshop of the 14th European Conference on Artificial Intelligence (ECAI 2000)

This paper describes Mo'K, a configurable workbench that supports the development of conceptual clustering methods for ontology building. Mo'K is intended to assist ontology developers in the exploratory process of defining the most suitable learning methods for a given task. To do so, it provides facilities for evaluation, comparison, characterization and elaboration of conceptual clustering methods. Also, the model underlying Mo'K permits a fine- grained definition of similarity measures and class construction operators, easing the tasks of method instantiation and configuration. This paper presents some experimental results that illustrate the suitability of the model to help characterize and assess the performance of different methods that learn semantic classes from parsed corpora.