A Framework for Developing Public Data-map Using Similarity between Meta-data and Graph: The Case of Public Data from Seoul

Abstract

The South Korean government is actively working to make data available to the public. However, as data from different departments is integrated and made accessible, efficient search algorithms for big data have become a major issue. This paper proposes a framework for developing a public data-map that uses metadata similarity and graph concepts to suggest ways to visualize and search related data. Additionally, to improve the performance of measuring similarity, we develop the domain-specific data pre-processing for public data and incorporate the step into the framework. To validate the framework, an empirical study was conducted using the case of the Seoul Metropolitan Government’s Big Data Division. The results show that this framework can significantly improve the usability of public data and facilitate its open access.