Context-aware mobile diary

Human memory is fallible. As a remedy to this problem, in the near future wearable devices will have the potential to automatically record every bit of information concerning a person and his/her environment continuously. This way, lifelog of a person can be generated. In this thesis, we present a mobile lifelogging application that captures daily experiences of a smartphone user and allows to retrieve them later. In this direction, we first introduce a system named Smartphone Tracker which allows us to conduct large-scale data collection studies using smartphones. Using this system, we investigate the sensing capabilities of smartphones through a real life data collection study with 22 participants. Second, we present a novel algorithm which provides semantic location awareness to mobile devices. Using this algorithm, we are able to detect the entry/departure times to/from semantically meaningful places a user visits, such as home, office, parents' home, etc. Finally, we present a mobile lifelogging application named Auto Diary which automatically records SMS messages, phone calls, weather conditions, ambient audio and visited locations. In addition, a retrieval functionality is offered which allows to retrieve past experiences via associative cues. For example, the cues from the statements ``It was a rainy Monday morning. I was in my office and received a phone call from my brother between 10 am and 12 am." can be used as query terms in the application to retrieve the experiences (e.g., audio recordings) captured in the described day. We envision that Auto Diary can be useful for people experiencing episodic memory impairment. It can also be useful for people who want to remember their past experiences with greater precision.