Article ID: 2020XBL0025
In this paper, we propose a real-time detection system for smartphone zombie (smombie) base on machine learning. The feature of this study is that it does not use cameras and external devices, which are often seen in studies focused on smombie. The proposed system uses four kinds of brightness, acceleration, gyro and motion sensors. We find the effectiveness of the proposed system by experiments for 10 examinees. We show it can detect smombie with 93 % accuracy.