Suitable Data Representation for Abnormal Pattern Detection in Smart Home Environment.

Development in the area of sensor technologies and subsequently applications within the Internet of Things allows the implementation of systems for continuous monitoring of human's vital parameters and daily activities. This development is welcomed since the population aged 65 and over is constantly increasing. Moreover, the number of seniors living alone is also increasing. The monitoring systems can contribute to the safety and security of elderly people and allow them to stay at home where they are used to live as long as possible. Application of various sensor types raises questions on the most suitable sensor data representation, not losing useful information, and also on the design of detection and evaluation algorithms. In the paper, we briefly present several existing approaches and explain why we decided to use the basic ideas coming from the area of behaviour informatics.