We are developing an eParticipation web platform based on Linked Open Data that targets regional communities in Japan. To increase transparency and public participation, we aim to utilize web contents related to target regions for sharing public concerns among citizens, government officials, and experts. We have designed a Linked Open Data set called SOCIA (Social Opinions and Concerns for Ideal Argumentation) to structure regional web contents (e.g. regional news articles, microblog posts, and minutes of city council meetings) and utilize them for eParticipation and concern assessment. The web contents are semi-automatically structured by our text mining system, Sophia, on the basis of regions and events extracted from news articles on the web. Minutes of city council meetings stored in SOCIA are annotated with discourse salience in order to visualize topic transitions in a meeting transcript. We also developed a prototype debate support system called citispe@k that uses SOCIA to help citizens share their concerns. Users can submit agendas, ideas, questions, and answers by referencing the structured regional information in SOCIA. Moreover, they can annotate SOCIA data with tags representing criteria for assessing concerns or utterance intentions.