報告題目：Privacy Leakage in Smart Devices Through Seemingly Innocuous Sensory Data
報告人：Dr. Yingshu Li
報告人簡介：Dr. Yingshu Li received her PhD and MS degrees from the Department of Computer Science and Engineering at University of Minnesota-Twin Cities. She received her BS degree from the Department of Computer Science and Engineering at Beijing Institute of Technology, China. Dr. Li is currently an Associate Professor in the Department of Computer Science at Georgia State University. Her research interests include Wireless Sensor Networks, Big Sensory Data Management, Privacy-aware Computing, Smart Cities, Social Networks, and Wireless Networking. Dr. Li is the recipient of an NSF CAREER Award. She has published more than 150 papers including 20 ACM/IEEE Transactions papers. Her publications have received almost 6000 citations and her h-index is 35. Dr. Li has served as an associate/guest editor for several prestigious journals including IEEE Transactions on Computers, ACM Transaction on Senor Networks. IEEE Internet of Things Journal, etc. Dr. Li was the steering committee chair for the international conference WASA and will be the program chair for COCOA 2019 and IPCCC 2018.
報告簡介：Smart devices and mobile apps are rolling out at swift speeds over the last decade, turning these devices into convenient and general-purpose computing platforms. However, smart devices also bring risks of privacy leakage. Sensory data collected from smart devices are important resources to nourish mobile services, and they are regarded as innocuous information which can be obtained without user permission and awareness. In this talk, two privacy issues caused by sensory data collected from smart devices will be addressed. The first one is location privacy leakage. The second one is deep learning based inference of private information. For the first issue, it is shown that only by using the data collected from the embedded sensors in smart devices instead of GPS data, a user's location information can be inferred with high accuracy. For the second issue, it is shown that the seemingly innocuous sensory data could help with infer users' tap positions on the screens of smart devices by employing some deep learning techniques. The tap stream profiles for each type of apps can then be derived so that a user's app usage habit can be accurately inferred. Furthermore, it is shown that users' app usage habits and passwords may be inferred with high accuracy.