Watch:
In this article, the authors shed light on the challenge of underutilizing the big data generated by smart cities from a machine learning perspective. In particular, they discuss the phenomenon of wasting unlabeled data and they argue that semi-supervision is a must for smart cities to address this challenge. Finally, they propose a three-level learning framework for smart cities that matches the hierarchical nature of big data generated by smart cities.In this article, the authors shed light on the challenge of underutilizing the big data generated by smart cities from a machine learning perspective. In particular, they discuss the phenomenon of wasting unlabeled data and they argue that semi-supervision is a must for smart cities to address this challenge. Finally, they propose a three-level learning framework for smart cities that matches the hierarchical nature of big data generated by smart cities.
Nowadays, many cities collect measurements from several sources of data about the environment, through IoT and other sensing devices that are placed around the city. However, accessing raw data on its own is not sufficient to meaningfully show the data being collected and to allow a range of users to explore what this data may mean for them. This paper presents a novel approach for visualizing urban data and has been implemented as a pilot in Newcastle-upon-Tyne, UK.