The domain ontology, which plays a significant role in knowledge-based systems, still needs the manual work of domain experts to be constructed currently. The main motivation of this paper is to provide a semi-automatic platform which can construct fairly comprehensive domain ontology from unstructured data. Firstly, a brief QA process is proposed to simplify the interaction with the domain experts. A novel algorithm MPVW, which extends from the classical algorithm TF-IDF, is proposed to extract the terminologies from domain documents. MPVW balanced more parameters and factors to evaluate the feature of terminologies. The 3-layers taxonomy and terminology hyponymy height provide sufficient guide and prompt for domain experts to construct ontology from terminologies. According to our approach we have developed ROCP, a rapid ontology construction platform which has been applied in the space debris mitigation domain. The experimental data indicates that ROCP has sufficient accuracy to extract terminologies. Meanwhile, it is effective to relieve the labor of domain experts to construct domain ontology.
CITATION STYLE
Zhao, C., Dong, C., & Zhang, X. (2018). ROCP: A rapid ontology construction platform from unstructured data. Data Science Journal, 17. https://doi.org/10.5334/dsj-2018-023
Mendeley helps you to discover research relevant for your work.