iMiner: Sub-room-level POI interaction detection for semantic location history construction

Sub-room-level location-based services are poised to impact people's daily lives in the near future. Sub-room-level semantic location history (SLH) construction is an essential component of such services. However, little in-depth research on constructing SLH in sub-room-level scenarios has been conducted. In this paper, we propose iMiner, an automatic sub-room-level SLH construction system. iMiner consists of an improved Kalman filter-based approach to obtain accurate location and heading estimation from inertial sensor readings of smartphones. The collected users' location and heading histories arc further mined to estimate areas of interaction (AOIs). AOT is a novel idea and a powerful tool for accurately embedding points of interest into geographic trajectories. A series of experiments were conducted in two large-scale sites. The results have demonstrated the effectiveness of the proposed system.