Search

Thanks to advances in Internet of Things and crowd-sensing, it is possible to collect vast amounts of urban data, to better understand how citizens interact with cities and, in turn, improve human well-being in urban environments. This is a scientifically challenging proposition, as it requires new methods to fuse objective (heterogeneous) data (e.g. people location trails and sensors data) with subjective (perceptual) data (e.g. the citizens’ quality of experience collected through feedback forms). When it comes to vast urban areas, collecting statistically significant data is a daunting task; thus new data-collection methods are required too. In this work, we turn to artificial intelligence (AI) to address these challenges, introducing a method whereby the objective, sensor data is analyzed in real-time to scope down the test matrix of the subjective questionnaires. In turn, subjective responses are parsed through AI models to extract further objective information. The outcome is an interactive data analysis framework for urban environments, which we put to test in the context of a citizens’ well-being project. In our pilot study, each new entry (objective or subjective) is parsed through the AI engine to determine which action maximizes the information gain. This translates into a particular question being fired at a specific moment and place, to a specific person. With our AI data collection method, we can reach statistical significance much faster, achieving (in our city-wide pilot study) a 41% acceleration factor and a 75% reduction in intrusiveness. Our study opens new avenues in urban science, with potential applications in urban planning, citizen’s well-being projects, and sociology, to mention but a few cases.

Recent developments in cloud computing and the Internet of Things have enabled smart environments, in terms of both monitoring and actuation. Unfortunately, this often results in unsustainable cloud-based solutions, whereby, in the interest of simplicity, a wealth of raw (unprocessed) data are pushed from sensor nodes to the cloud. Herein, we advocate the use of machine learning at sensor nodes to perform essential data-cleaning operations, to avoid the transmission of corrupted (often unusable) data to the cloud. Starting from a public pollution dataset, we investigate how two machine learning techniques (kNN and missForest) may be embedded on Raspberry Pi to perform data imputation, without impacting the data collection process. Our experimental results demonstrate the accuracy and computational efficiency of edge-learning methods for filling in missing data values in corrupted data series. We find that kNN and missForest correctly impute up to 40% of randomly distributed missing values, with a density distribution of values that is indistinguishable from the benchmark. We also show a trade-off analysis for the case of bursty missing values, with recoverable blocks of up to 100 samples. Computation times are shorter than sampling periods, allowing for data imputation at the edge in a timely manner.