Toward automatic assessment of the categorization structure of open data portals

Abstract

Governments worldwide have been releasing their owned data recently for public usage and arising lots of novel applications and services. Issues on open data were also intensively discussed from researchers and practitioners. One of the key issues in adopting open data is the accessibility of the data, which are generally collectively provided in open data portals. Open data portals categorize open datasets according to their domains, providers, formats, and other properties for better accessibility of the data. However, these portals did not follow a conforming standard in establishing their categorization structures. In this work, we try to assess the goodness of categorization structures of open data portals automatically by investigating the coherence of the datasets in the same category. The detailed methodology is described but preliminary experiments on Taiwan’s open data portals are still undergoing.