Construction of Urban Problem LOD using Crowdsourcing

Municipalities in Japan have various urban problems such as traffic accidents, illegally parked bicycles, and noise pollution. However, using these data to solve urban problems is difficult, as these data are not structurally constructed. Hence, we aim to construct the Linked Data set that will facilitate the solving of urban problems. In this paper, we propose a method for semi-automatic construction of Linked Data with the causality of urban problems, based on Web pages and open government data. Specifically, we extracted causal relations using natural language processing and crowdsourcing to include problem causality in the Linked Data. Then, we provided an example query to confirm the relationships between several problems. Finally, we discussed our crowdsourcing task design for extracting urban problem causality.