Evidence-based decision making with data-driven insights with big and small data
Floods of data about the movement of individuals, from infrastructure sensors, transport smart cards, and social networks, already provide large corporations and government agencies with data about individual movements. Movement patterns and mobility behaviours are vital information in many applications, for example, in the health and ageing domain, transport and urban monitoring, emergency response applications, crowd monitoring, and indoor space and facility management. However, some of these data may not be completely available to individuals and private and small medium enterprises and local councils and agencies. On the other side of the coin, there are small data from individuals carrying sensors, wearables, and smartphones, which can provide insights, not only to the routine behaviours and the anomalies, but also into the contexts and situations of urban areas around the individuals. These data can also potentially provide richer semantics or annotations into individuals’ or groups’ behaviours. If aggregated, in a privacy-preserving manner, both of these big and small data can build a rich fine-grained multi-resolution depiction of dynamic urban patterns and provide insights into the what, when, where, and why certain phenomena arises and potentially how to address them.
As a computer scientist researcher, specialising in mobile and ubiquitous computing and urban computing research areas, I have had the opportunities in dealing with various big and small data that may have come from smartphones, Wi-Fi, Bluetooth Beacons, road sensors, parking sensors, and other types of infrastructure sensors, across various projects. The opportunities that arise from the big and small data also come with challenges inherent with these data — the noise, uncertainty, incompleteness, and sparsity of data, which must be dealt with before we can even start extracting interesting insights from the data and getting them used by the data stakeholders. These are the challenges faced by my CRUISE research team in RMIT University, and we have generated new frameworks, methods, and techniques to deal with these challenges, and also predictive models for monitoring urban dynamics and human mobility in several outdoors and indoors case studies.
I am passionate about solving urban mobility and transportation issues, and also population health and ageing issues, using data science and urban computing lens, and also using user-centred design approaches. I was a postdoc in an architecture and design lab. My career background makes me a versatile researcher with multidisciplinary strength, with knowledge of works in ubiquitous and pervasive computing, design and architecture, and also in emerging domain areas of urban computing. As an academic from RMIT University, I teach User Centred Design course, and I always teach my students to involve users right from the outsets. There needs to be user-centric design approaches in order to get users to participate and contribute their data in a crowdsensing project, an area that needs further explorations if we are to obtain more useful data for evidence-driven decision making for urban sustainability.
It is a privilege for me to be one of the 1,000 global talents to participate in UNLEASH Lab 2017 aimed at solving UN’s Sustainable Development Goals. I am assigned to Urban Sustainability theme, a theme that I am passionate about. I believe when the right connections are in place, the possibilities are limitless. What UNLEASH does for us is to facilitate that connections in such a grand scale that the impossibilities could happen. My hope is that as we put the pieces together, we can build solutions that people (and citizens of the world) would be willing to be a part of. And where else in the world this should be held other than Denmark, where LEGO was invented! I thought it is a very nicely planned coincidence.
