A data-driven approach for the generation of a liveability index based on the UBER API
Data Science, Uber, Liveability Indicators, Urbanization, Sustainable Urban Development
One of the global dilemmas concerns the accelerated demographic transition to urban areas in the last decades. Therefore, promoting sustainable urban development to accommodate that population growth is extremely important. Under those circumstances, the concept of liveability arises, being widely discussed in the New Urban Agenda (NUA) adopted by the United Nations in 2016. NUA defines policies to promote the Sustainable Development Goals (SDGs) consolidation, particularly Goal 11 focused on a pro-urban future. To achieve those goals, it is required to use indicators to supervise SDGs implementation, as liveability indicators for example. However, there are issues related to current data unavailability, poor quality and aggregation, making the SDGs monitoring difficult. Considering the described scenario, this work proposes a liveability index based on Uber estimated time of arrival and estimated cost of service. A data science approach was conducted over data sourced from the Brazilian city of Natal (RN), covering from data acquisition and cleaning to Exploratory Data Analysis (EDA), and was followed by the multivariate index creation using statistical techniques. EDA results show how the Uber service reacts to weather variations, festivals and other events, as well as the correlations of the data with social indicators. Finally, the proposed index can be applied in sustainable development monitoring and can still be used to draw parallels between different cities.