Event Classification Using Adaptive Cluster-Based Ensemble Learning of Streaming Sensor Data
Sensor data stream mining methods have recently brought significant attention to smart homes research. Through the use of sliding windows on the streaming sensor data, activities can be recognized through the sensor events. However, it remains a challenge to attain real-time activity recognition from the online streaming sensor data. This paper proposes a new event classification method called Adaptive Cluster-Based Ensemble Learning of Streaming sensor data (ACBEstreaming). It contains desirable features such as adaptively windowing sensor events, detecting relevant sensor events using a time decay function, preserving past sensor information in its current window, and forming online clusters of streaming sensor data. The proposed approach improves the representation of streaming sensor-events, learns and recognizes activities in an on-line fashion. Experiments conducted using a real-world smart home dataset for activity recognition have achieved better results than the current approaches.
KeywordsOn-line learning Streaming Activity recognition Smart home
- 4.Rashidi, P.: Stream sequence mining for human activity discovery. In: Sukthankar, G., Geib, C., Bui, H., Pynadath, D., Goldman, R.P. (eds.) Plan, Activity, and Intent Recognition. Elsevier (2014)Google Scholar
- 5.Rashidi, T., Cook, D.J.: Adapting to resident preferences in smart environments. In: Proceedings of the AAAI Workshop on Advances in Preference Handling (2008)Google Scholar
- 9.van Kasteren, T., Noulas, A., Englebienne, G., Kröse, B.: Accurate activity recognition in a home setting. In: Proceedings of the 10th International Conference on Ubiquitous Computing, UbiComp 2008, pp. 1–9. ACM, New York (2008)Google Scholar
- 12.Zhou, Z.-H.: Ensemble Methods: Foundations and Algorithms, 1st edn. Chapman & Hall/CRC, Boca Raton (2012) Google Scholar
- 14.Bifet, A., Holmes, G., Kirkby, R., Pfahringer, B.: MOA: massive online analysis. J. Mach. Learning Res. 11, 1601–1604 (2010)Google Scholar