SOCIA: Linked Open Data of Context behind Local Concerns for Supporting Public Participation

Abstract

To address public concerns that threat the sustainability of localsocieties, supporting public participation by sharing the backgroundcontext behind these concerns is essentially important. We designed aSOCIA ontology, which was a linked data model, for sharing contextbehind local concerns with two approaches: (1) structuring Web newsarticles and microblogs about local concerns on the basis ofgeographical regions and events that were referred to by content, and(2) structuring public issues and their solutions as public goals. Wemoreover built a SOCIA dataset, which was a linked open dataset, on thebasis of the SOCIA ontology. Web news articles and microblogs related tolocal concerns were semi-automatically gathered and structured. Publicissues and goals were manually extracted from Web content related torevitalization from the Great East Japan Earthquake. Towards moreaccurate extraction of public concerns, we investigated featureexpressions for extracting public concerns from microblogs written inJapanese. To address a technical issue about sample selection bias inour microblog corpus, we formulated a metric in mining featureexpressions, i.e., bias-penalized information gain (BPIG). Furthermore,we developed a prototype of a public debate support system that utilizedthe SOCIA dataset and formulated the similarity between public goals fora goal matching service to facilitate collaboration.