Structuring of lifelog captured with multiple sensors by using neural network
Masayuki Ono, Kunihiro Nishimura, Tomohiro Tanikawa, Michitaka Hirose · 2010
The data related to our life experiences is called lifelog and it can easily be collected these days. Although lifelog research has been conducted for a long time, its application is still under development. In our research, we aimed at structuring of a lifelog with multiple sensors for the purpose of the application such as tools to recall memories. We actually captured a lifelog data with a device that has multiple sensors, and we estimated the events, i.e., the participants' activities. Then, the system correctly estimated events 70.4% of the time. Moreover, we efficiently showed the estimated results on the viewer we created. Finally, we verified the effectiveness of the research in experiment with two participants.