Public participation is crucial in the process of urban governance in smart-city initiatives to enable urban planners and policy makers to take account of the real public needs. Our study aims to develop an analytical framework using citizen-centred qualitative data to analyse urban problems and identify the areas most needed for urban governance. Taking a Chinese megacity as the study area, we first utilise a web-crawling tool to retrieve public comments from an online comment board and employ the Baidu Application Programming Interfaces and a qualitative content analysis for data reclassification. We then analyse the urban problems reflected by negative comments in terms of their statistical and spatial distribution, and the associative factors to explain their formation. Our findings show that urban problems are dominantly related to construction and housing, and most frequently appear in industry-oriented areas and newly-developed economic development zones on the urban fringe, where the reconciling of government-centered governance and private governance by real estate developers and property management companies are most needed. Areas with higher land price and a higher proportion of aged population tend to have less urban problems, while various types of civil facilities affect the prevalence of urban problems differently.

Policy induced decline of human mobility has been recognised to be effective in controlling the COVID-19 spread especially in the initial stage of the outbreak, although the relationship among mobility, policy implementation, and virus spread remains contentious. Coupling data of confirmed COVID-19 cases with Google mobility data in Australia, we present a state-level empirical study to: 1) inspect the temporal variation of COVID-19 spread and the change of mobility adherent to social restriction policies; 2) examine the extent that different types of mobility are associated with the COVID-19 spread in eight Australian states/territories; and 3) analyse the time-lag effect of mobility restriction on the COVID-19 spread. We find that social restriction policies implemented in the early stage of the pandemic controlled the COVID-19 spread effectively; the restriction of human mobility has a time-lag effect on growth rates, and the strength of the mobility-spread correlation increases up to seven days after policy implementation but decreases afterwards. The association between mobility and COVID-19 spread varies across space and time, and subjects to the types of mobility. Thus, it is important for governments to consider the degree to which lockdown conditions can be eased by accounting for this dynamic mobility-spread relationship.

Studies on human mobility have a long history with increasingly strong connections to multi-disciplines across social science, environmental science, information and technology, computer science, engineering, and health science. However, what is lacking in the current research is a summary of studies on human mobility to identify the evolutional pathway and future research directions. To address this gap, we conduct a systematic review of human mobility-related studies published from 1990 to 2020. Drawing on the selected publications retrieved from the Web of Science, we conduct a bibliometric analysis and network visualisation by CiteSpace and VOSviewer on publication years and numbers, authors and their countries and afflictions, citations, topics, abstracts, keyword, and journals. Our findings show that human mobility-related studies have become increasingly interdisciplinary and multi-dimensional, enhanced by the involvement of multi-source big data, the development of technologies, the innovation of modelling approaches, and the novel applications in various areas. We also summarise four future research directions proposed by top cited authors and mobility studies, in terms of data sources, modelling methods, applications, and technologies. We advocate in-depth research on human mobility to address real-world problems and contribute to social good as promising future orientations through integrating multi-source big data and advanced modelling methods facilitated by artificial intelligence, and machine and deep learning.