… can highly limit gender recognition (from 89.37% using all the features to 64.38% after
applying the MOEA algorithm) while only reducing the accuracy of activity recognition by 5.45% …

… Balancing activity recognition and privacy preservation with multi-objective evolutionary
algorithm. In Proceedings of the 7th EAI International Conference on Smart Objects and …

… For example, human activity recognition [1… 2], age group detection [3], health monitoring and
diagnosis [4], … We provide an accelerometer-based recognizer that works in the table setting …

… Weiss and Lockhart emphasize that accelerometer-based gender recognition can work
independently of a subject’s weight and height [56]. Even acoustic vibrations caused by a person’…

… We propose a new side-channel attack on smartphone speech privacy, developed through
a new accelerometer-based recognizer of words that works in the table setting scenario. …

… We show that, on gesture and activity recognition tasks, we can prevent inference of … To
evaluate the RAE, we use four benchmark datasets of gesture and activity recognition including …

… in human recognition and human activity recognition. No … -of-the-art works for privacy-preserving
human activity sensing. … works in privacy-preserving human activity sensing considering …

… However, the major challenge of accelerometer-based gait recognition is the continuous
sampling accelerometer will quickly drain the battery of wearable devices. Energy constraint …

… privacy issues and corresponding approaches in human recognition and human activity
recognition… of-the-art works for privacy-preserving human activity sensing. Therefore, we expand …

… this setup and explore accelerometer-based speech recognition and reconstruction … activity
recognition, 50 Hz is also more than sufficient since the frequencies of most human activities …