Relation extraction from documents for the automatic construction of ontologies

The Semantic Web relies on domain ontologies that structure underlying data enabling comprehensive and transportable machine understanding. It takes so much time and efforts to construct domain ontologies because these ontologies have to be manually made by domain experts and knowledge engineers. To solve this problem, there have been some researches to semi-automatically construct ontologies. In this paper, we propose a hybrid method to extract relations from domain documents which combines a named relation approach and an unnamed relation approach. Our named relation approach is based on the Snowball system. We add the generalized pattern method to their methods. In our unnamed relation approach, we extract unnamed relations using association rules and clustering method. We also recommend candidate names of unnamed relations. We experiment and evaluate the proposed method by using Ziff documents set offered by TREC.