Unsupervised Habitual Activity Detection in Accelerometer Data
The activity of the user is one example of context information which can help computer applications respond better to the needs of the user in a seamless manner based on the situation without needing explicit instruction. With potential applications in many fields such as health-care, assisted living and sports, there has been considerable interest and work done in the area of activity recognition. Currently, these works have resulted in various successful approaches capable of recognizing common basic activities such as walking, sitting, standing and lying, mostly through supervised learning. However, supervised learning approach would be limited in that it requires labeled data for prior learning. It would be difficult to provide sufficient amounts of labeled data that is representative of free-living activities. To address these limitations, this research proposes motif discovery as an unsupervised activity recognition approach. Habitual activities would be detected by finding motifs, similar repeating subsequences within the collected accelerometer data. A 3D accelerometer sensor worn on the dominant arm is used to record, model and recognize different activities of daily living. The raw accelerometer data is then processed and discretized in order to perform motif discovery. Results have shown motif discovery to increase the performance in varying degrees (5–19%) depending on the discretization technique used.
- 1.Ajmera, J., H. Bourlard, I. Lapidot, and I.A. McCowan. 2002. Unknown-multiple speaker clustering using hmm. IDIAP: Technical Representative.Google Scholar
- 3.Bao, L., and S.S. Intille. 2004. Activity recognition from user-annotated acceleration data. In: Pervasive computing, 1–17. Springer.Google Scholar
- 7.Fuad, M.M.M., and P.F. Marteau. 2013. Towards a faster symbolic aggregate approximation method. arXiv preprint arXiv:1301.5871.
- 9.Gusfield, D. 1997. Algorithms on strings, trees and sequences: Computer science and computational biology. Cambridge University Press.Google Scholar
- 10.Hamid, R., and S. Maddi, A. Bobick, I. Essa. 2006. Unsupervised analysis of activity sequences using event-motifs. In: Proceedings of the 4th ACM international workshop on Video surveillance and sensor networks, 71–78. ACM.Google Scholar
- 12.Kasabach, C., C. Pacione, M. Des, and A. Teller, D. Andre. 2002. Why the upper arm? Factors contributing to the design of an accurate and comfortable, wearable body monitor. In Whitepaper, Bodymedia, Inc. Citeseer.Google Scholar
- 13.Lee, M.S., J.G. Lim, K.R. Park, and D.S. Kwon. 2009. Unsupervised clustering for abnormality detection based on the tri-axial accelerometer. ICCAS-SICE 2009: 134–137.Google Scholar
- 14.Lin, J., E. Keogh, and S. Lonardi, P. Patel. 2002. Finding motifs in time series. In Proceedings of the 2nd workshop on temporal data mining, 53–68.Google Scholar
- 15.Logan, B., J. Healey, M. Philipose, and E.M. Tapia, S. Intille. 2007. A long-term evaluation of sensing modalities for activity recognition. In UbiComp 2007: Ubiquitous computing, 483–500. Springer.Google Scholar
- 17.Mueen, A., and E. Keogh. 2010. Online discovery and maintenance of time series motifs. In: Proceedings of the 16th ACM SIGKDD international conference on knowledge discovery and data mining, 1089–1098. ACM.Google Scholar
- 18.Nguyen, A., and D. Moore, I. McCowan. 2007. Unsupervised clustering of free-living human activities using ambulatory accelerometry. In Engineering in medicine and biology society, 2007. EMBS 2007. 29th annual international conference of the IEEE, 4895–4898, IEEE.Google Scholar
- 20.Siirtola, P., P. Laurinen, E. Haapalainen, and J. Roning, H. Kinnunen. 2009. Clustering-based activity classification with a wrist-worn accelerometer using basic features. In Computational intelligence and data mining, 2009. CIDM’09. IEEE symposium on IEEE, 95–100.Google Scholar
- 22.Trabelsi, D., S. Mohammed, and Y. Amirat, L. Oukhellou. 2012. Activity recognition us-ing body mounted sensors: An unsupervised learning based approach. In Neural networks (IJCNN), The 2012 international joint conference on IEEE, 1–7.Google Scholar