Approaches to Assessing Public Concerns: Building Linked Data for Public Goals and Criteria Extracted from Textual Content
白松俊, 大囿忠親, 新谷虎松

アブストラクト

The importance of public involvement in Japanese regional societies is increasing because they currently face complicated and ongoing social issues due to the post-maturity stage of these societies. Since citizens who have beneficial awareness or knowledge are not always experts on relevant social issues, assessing and sharing public concerns are needed to reduce barriers to public participation. We propose two approaches to assess public concerns. The first is building a linked open data set by extracting public goals for a specific social issue aimed at by citizens or agents from articles or public opinions. This paper deals with hierarchical goals and subgoals for recovery and revitalization from the Great East Japan Earthquake manually extracted from related articles. The data set can be used for developing services to match citizens and agents who aim at similar goals to facilitate collaboration. The second approach is building a linked data set by extracting assessment criteria for a specific social issue from public opinions. This paper deals with candidate terms that potentially represent such criteria for a specific public project automatically extracted from clusters of citizens' opinions. The data set can be used as evidence for policy-making about the target project.