Dissecting urban noises from heterogeneous geo-social media and sensor data

Geo-social media services, such as Foursquare and Flickr, provide rich data that sensors various urban activities of hu-man beings from geographical, mobility, visual, and social aspects. While noise pollution in modern cities is getting worse and sound sensors are sparse and costly, it is highly demanded to infer and analyze the noise at any region in urban areas. In this paper, we aim to leverage heterogeneous geo-social sensor data on Foursquare, Flickr, and Gowalla, to dissect urban noises for every regions in a city. Using NYC 311 noise complaint records as the approximation of urban noises generated by regions, we propose a novel unsuper-vised framework that integrates the extracted geographical, mobility, visual, and social features to infer the noise compo-sition for regions and time intervals of interest in NYC. Ex-perimental results show that our system can achieve promis-ing results with substantially few training data, compared to state-of-The-Art methods.