Enhancing the Privacy and Computability of Location-Sensitive Data for Context Authentication
share
This paper proposes a new privacy-enhancing, context-aware user authentication system, ConSec, which uses a transformation of general location-sensitive data, such as GPS location, barometric altitude and noise levels, collected from the user's device, into a representation based on Locality-Sensitive-Hashing (LSH). This enables numerical computation on the hashed data, which is used by a machine learning classifier to model user behaviour for authentication. We present how ConSec supports learning from categorical as well as numerical data, while addressing a multitude of on-device and network-based threats. ConSec is implemented for the Android platform and an extensive evaluation is presented using data collected in the field from 35 users. Experimental results are promising, with a small equal error rate of 2.5 privacy analysis of the proposed scheme before, lastly, concluding with some future areas of research.
READ FULL TEXT