Classification of Wikipedia pages using Wikipedia category ontology

Abstract

Since a category of Wikipedia contains information useful for classifying a Wikipedia pages. A classification method using this category information has been proposed.However, since various types of information are mixed in the Wikipedia category, there is a problem that a category that is not appropriate to be assigned at the same time is added.In this research, we classify more appropriate classes for pages with multiple categories using Wikipedia category ontology, which defines the types of Wikipedia categories and their parent-child relationships. By using this method, we rank the adequacy of the page class belonging to a plurality of categories. To confirm effectiveness of this method, we compar with the classification of SHINRA : Wikipedia Structuring Project.