Large scale human mobility data can be collected from mobile phones, car navigation systems, location-based applications, social media, Wi-Fi, and traffic cameras. Turning such raw data into knowledge can provide insights in social science, urban problems, and prevention health, and can also benefit applications in transportation, advertisement targeting, and urban planning. In this talk, I would like to share our recent research work in using innovative data mining techniques on human mobility data. First, I would like to present our new semantic annotation techniques on mobility data, which turn raw mobility data into semantic activity space by associating them with surrounding contexts. Next, I will share our recent discovery on connectivity of urban regions through the mobility flows. We propose region representation learning via flows and demonstrate the use of such representations in predicting crimes and region properties. Lastly, I will introduce our new spatial-temporal deep learning models that demonstrate superior performance in predicting taxi demands and traffic volume.
Speaker: Jessie Li, Penn State
Contact:Website: Click to Visit
Save this Event:iCalendar
Windows Live Calendar