Hristijan Gjoreski
ABSTRACT
We present a novel unsupervised approach, UnADevs, for discovering activity clusters corresponding to periodic and stationary activities in streaming sensor data. Such activities usually last for some time, which is exploited by our method; it includes mechanisms to regulate sensitivity to brief outliers and can discover multiple clusters overlapping in time to better deal with deviations from nominal behaviour. The method was evaluated on two activity datasets containing large number of activities (14 and 33 respectively) against online agglomerative clustering and DBSCAN. In a multi-criteria evaluation, our approach achieved significantly better performance on majority of the measures, with the advantages that: (i) it does not require to specify the number of clusters beforehand (it is open ended); (ii) it is online and can find clusters in real time; (iii) it has constant time complexity; (iv) and it is memory efficient as it does not keep the data samples in memory. Overall, it has managed to discover 616 of the total 717 activities. Because it discovers clusters of activities in real time, it is ideal to work alongside an active learning system.
- A. Bulling, U. Blanke, and B. Schiele, "A tutorial on human activity recognition using body-worn inertial sensors," ACM Comput. Surv., vol. 1, no. June, pp. 1--33, 2014. Google Scholar
Digital Library
- D. J. Cook, N. C. Krishnan, and P. Rashidi, "Activity Discovery and Activity Recognition: A New Partnership," Cybern. IEEE Trans., vol. 43, no. 3, pp. 820--828, 2013.Google ScholarCross Ref
- Y. Kwon, K. Kang, and C. Bae, "Unsupervised learning for human activity recognition using smartphone sensors," Expert Syst. Appl., vol. 41, no. 14, pp. 6067--6074, 2014.Google ScholarCross Ref
- E. Berlin and K. Van Laerhoven, "Detecting leisure activities with dense motif discovery," Proc. 2012 ACM Conf. Ubiquitous Comput. - UbiComp '12, p. 250, 2012. Google ScholarDigital Library
- N. Begum and E. Keogh, "Rare Time Series Motif Discovery from Unbounded Streams," Vldb, vol. 8, no. 2, pp. 149--160, 2014. Google ScholarDigital Library
- D. Minnen, T. Starner, M. Essa, and C. Isbell, "Discovering characteristic actions from on-body sensor data," Proc. - Int. Symp. Wearable Comput. ISWC, pp. 11--20, 2007.Google Scholar
- T. Maekawa, D. Nakai, K. Ohara, and Y. Namioka, "Toward practical factory activity recognition: Unsupervised understanding of repetitive assembly work in a factory," UbiComp 2016, pp. 1--12, 2016. Google ScholarDigital Library
- J. Seiter, W. C. Chiu, M. Fritz, O. Amft, and G. Tröster, "Joint segmentation and activity discovery using semantic and temporal priors," 2015 IEEE Int. Conf. Pervasive Comput. Commun. PerCom 2015, pp. 71--78, 2015.Google Scholar
- T. Huynh, M. Fritz, and B. Schiele, "Discovery of activity patterns using topic models," Proc. 10th Int. Conf. Ubiquitous Comput. (UbiComp '08), pp. 10--19, 2008. Google ScholarDigital Library
- T. Miu, D. Roggen, P. Missier, and T. Plötz, "On strategies for budget-based online annotation in human activity recognition," UbiComp 2014 - Adjun. Proc. 2014 ACM Int. Jt. Conf. Pervasive Ubiquitous Comput., pp. 767--776, 2014. Google ScholarDigital Library
- M. Stikic, K. Van Laerhoven, and B. Schiele, "Exploring semi-supervised and active learning for activity recognition," Wearable Comput. 2008. ISWC 2008. 12th IEEE Int. Symp., pp. 81--88, 2008. Google ScholarDigital Library
- "Withings Activity Recognition Smartwatch." {Online}. https://health.nokia.com/blog/2016/12/21/introducing-our-new-activity-recognition-feature. {Accessed: 10-Apr-2017}.Google Scholar
- I. D. Guedalia, M. London, and M. Werman, "An On-line agglomerative clustering method for nonstationary data.," Neural Comput., vol. 11, no. 2, pp. 521--540, 1999. Google ScholarDigital Library
- D. Zhang, S. Chen, and K. Tan, "Improving the robustness of 'online agglomerative clustering method' based on kernel-induce distance measures," Neural Process. Lett., vol. 21, no. 1, pp. 45--51, 2005. Google ScholarDigital Library
- H. Gjoreski, B. Kaluža, M. Gams, R. Milić, and M. Luštrek, "Context-based ensemble method for human energy expenditure estimation," Appl. Soft Comput., vol. 37, pp. 960--970, 2015. Google ScholarDigital Library
- M. Damas, O. Amft, A. M. Toth, O. Baños, H. Pomares, and M. A. Tóth, "A benchmark dataset to evaluate sensor displacement in activity recognition," ACM Conf. Ubiquitous Comput. - UbiComp '12, vol. 16, p. 1026, 2012. Google ScholarDigital Library
- M. Ester, H. P. Kriegel, J. Sander, and X. Xu, "A Density-Based Algorithm for Discovering Clusters in Large Spatial Databases with Noise," Proc. 2nd Int. Conf. Knowl. Discov. Data Min., pp. 226--231, 1996. Google ScholarDigital Library
- P. Siirtola, P. Laurinen, E. Haapalainen, J. Röning, and H. Kinnunen, "Clustering-based activity classification with a wrist-worn accelerometer using basic features," IEEE Symp. Comput. Intell. Data Min., pp. 95--100, 2009.Google Scholar
- S. Kisilevieh, F. Mansmann, M. Nanni, and S. Rinzivillo, "Spatio-temporal clustering," Data Min. Knowl. Discov. Handb., pp. 855--874, 2010.Google Scholar
- D. Birant and A. Kut, "ST-DBSCAN: An algorithm for clustering spatial-temporal data," Data Knowl. Eng., vol. 60, no. 1, pp. 208--221, 2007. Google ScholarDigital Library
- F. Zhou, S. Member, F. De Torre, and J. K. Hodgins, "Hierarchical Aligned Cluster Analysis for Temporal Clustering of Human Motion," IEEE Trans. Pattern Anal. Mach. Intell., vol. 35, no. 3, pp. 582--596, 2013. Google ScholarDigital Library
- J. Gong, P. Asare, Y. Qi, and J. Lach, "Piecewise Linear Dynamical Model for Action Clustering from Real-World Deployments of Inertial Body Sensors," IEEE Trans. Affect. Comput., vol. 7, no. 3, pp. 231--242, 2016. Google ScholarDigital Library
- B. Cvetković, R. Milić, and M. Luštrek, "Estimating Energy Expenditure with Multiple Models using Different Wearable Sensors," IEEE J. Biomed. Heal. informatics, vol. 20, 2016.Google Scholar
- M. D. R. Burkard and S. Martello, "Assignment Problems.," Soc. Ind. Appl. Math. SIAM, 2009. Google ScholarDigital Library
- H. Gjoreski et al., "Competitive live evaluations of activity-recognition systems," IEEE Pervasive Comput., vol. 14, no. 1, pp. 70--77, 2015.Google ScholarDigital Library
Index Terms
- Unsupervised online activity discovery using temporal behaviour assumption
Recommendations
Using wearable activity type detection to improve physical activity energy expenditure estimation
UbiComp '10: Proceedings of the 12th ACM international conference on Ubiquitous computingAccurate, real-time measurement of energy expended during everyday activities would enable development of novel health monitoring and wellness technologies. A technique using three miniature wearable accelerometers is presented that improves upon state-...
Recognition of Activities of Daily Living from Topic Model
Research in ubiquitous and pervasive technologies have made it possible to recognise activities of daily living through non-intrusive sensors. The data captured from these sensors are required to be classified using various machine learning or knowledge ...
Smartphone-based monitoring system for activities of daily living for elderly people and their relatives etc.
UbiComp '13 Adjunct: Proceedings of the 2013 ACM conference on Pervasive and ubiquitous computing adjunct publicationWe developed a smartphone-based monitoring system to allay the anxiety of elderly people and that of their relatives, friends and caregivers by unobtrusively monitoring an elderly person's activities of daily living. A smartphone of the elderly person ...
Comments