Urban sustainability management: A deep learning perspective

• Published in: Sustainable cities and society 2017-04, Vol.30, p.1-17
• Main Authors: Madu, Christian N., Kuei, Chu-hua, Lee, Picheng
• Format: Article
• Language: English
• Subjects: City carbon disclosure project (CDP); Deep learning; Formal concept analyses (FCA); Qualitative data mining; Urban sustainability management
• ISSN: 2210-6707; 2210-6715
• DOI: 10.1016/j.scs.2016.12.012

•This paper uses formal concept analyses (FCA) to identify focal points of the urban sustainability development initiatives.•A “deep” learning perspective is used to evaluate textual data obtained from City Carbon Disclosure Project (CDP).•Our focus is on three continents namely Europe, Asia, and North America.•Our empirical models show that the transportation sector is the focal point to reduce emissions in all the three continents.•No trend was observed with respect to the methodologies and guidelines applied. This paper uses formal concept analyses (FCA) and qualitative data points obtained from City Carbon Disclosure Project (CDP) to identify expected economic opportunities, the types of urban sustainability development incentives, emissions reduction activities, and methodologies/guidelines adopted for the on-going implementation of the urban sustainability development initiatives. Our focus is on three continents namely Europe, Asia, and North America. A “deep” learning perspective is used to evaluate textual data with depth of up to four layers. Association rules and concept lattice generation functions of FCA are employed and applied to support the learning process. Our empirical models show that the transportation sector is the focal point to reduce emissions in all the three continents. No trend was observed with respect to the methodologies and guidelines applied. There is a need to work interactively with the four layers of deep learning to establish new rules and guidelines for achieving reduction in emissions and urban sustainability transformations.