Citizen-centered big data analysis-driven governance intelligence framework for smart cities

Sensors and systems within rapidly expanding smart cities produce citizen-centered big data which have potential value to support citizen-centered urban governance decision-making. There exists a wealth of extant conceptual studies, however, further operational studies are needed to establish a specific path towards implementation of such data to governance decision-making with analytical algorithms that are appropriate for each step of the path. This paper proposes a framework for the use of citizen-centered big data analysis to drive governance intelligence in smart cities from two perspectives: urban governance issues and data-analysis algorithms. The framework consists of three layers: 1) A data-merging layer, which builds a citizen-centered panoramic data set for each citizen by merging citizen-related big data from multiple sources in collaborative urban governance via similarity calculation and conflict resolution; 2) a knowledge-discovery layer, which plots the citizen profile and citizen persona at both individual and group levels in terms of urban public service delivery and citizen participation via simple statistical analysis techniques, machine learning, and econometrics methods; and 3) a decision-making layer, which uses ontology models to standardize urban governance-related attributes, personas, and associations to support governance decision-making via data mining and Bayesian Net techniques. Finally, the proposed framework is validated in a case study on blood donation governance in China. This research highlights the value of citizen-centered big data, pushes data-to-decision research from conceptual to operational, synthesizes previously published frameworks for citizen-centered big data analysis in smart cities, and enhances the mutual supplement cross multiple disciplinaries.