Ontology-Based Social Media Analysis for Urban Planning

This paper reports a social media analysis study proposed by the Beijing Municipal Institute of Urban Planning and Design. The purpose is to explore techniques that can help the urban planning administrations to improve the social sensing and social perception abilities under the evolving data and technology environments. A framework integrating a comprehensive set of text mining algorithms is presented to conduct topic modeling, text clustering, event evolution detection, sentiment analysis, opinion mining, and information extraction on user-generated contents in Chinese social media. A domain ontology of Beijing urban planning is constructed to facilitate the text mining processes. Evaluations on two large, real-world datasets composed of microblogs and WeChat articles about the residential community and school education in Beijing demonstrate the effectiveness of our framework. The study illustrates the power of combining machine learning with knowledge-based, semantic approaches in analyzing social media for the domain of interest.