Context-aware applications using personal sensors
Abstract
We describe a framework based on the multi-agent paradigm to support context aware applications. This application is based on two agents classes, the context agent and personal agent. The first one deals with the sensor data and delivers context analysis, the second one uses user profile and the data received from context agents to provide personal and environmental aware applications to the end-user. Our framework focuses on knowledge exchanges, using ontology based representations. We used the concept of virtual knowledge communities to improve the knowledge management capabilities within the multi agent systems. In the second part of the paper, we present an application called 'Wake me up!' using accelerometer and mobile devices, to deliver personalized services to Tokyo's subway users.