Papers
Topics
Authors
Recent
Gemini 2.5 Flash
Gemini 2.5 Flash
149 tokens/sec
GPT-4o
7 tokens/sec
Gemini 2.5 Pro Pro
45 tokens/sec
o3 Pro
4 tokens/sec
GPT-4.1 Pro
38 tokens/sec
DeepSeek R1 via Azure Pro
28 tokens/sec
2000 character limit reached

Unsupervised Detection of Behavioural Drifts with Dynamic Clustering and Trajectory Analysis (2302.06228v2)

Published 13 Feb 2023 in cs.LG

Abstract: Real-time monitoring of human behaviours, especially in e-Health applications, has been an active area of research in the past decades. On top of IoT-based sensing environments, anomaly detection algorithms have been proposed for the early detection of abnormalities. Gradual change procedures, commonly referred to as drift anomalies, have received much less attention in the literature because they represent a much more challenging scenario than sudden temporary changes (point anomalies). In this paper, we propose, for the first time, a fully unsupervised real-time drift detection algorithm named DynAmo, which can identify drift periods as they are happening. DynAmo comprises a dynamic clustering component to capture the overall trends of monitored behaviours and a trajectory generation component, which extracts features from the densest cluster centroids. Finally, we apply an ensemble of divergence tests on sliding reference and detection windows to detect drift periods in the behavioural sequence.

Definition Search Book Streamline Icon: https://streamlinehq.com
References (53)
  1. J. Prochaska and W. Velicer, “The transtheoretical model of health behavior change,” American journal of health promotion : AJHP, vol. 12, pp. 38–48, 09 1997.
  2. S. Kincaid, “Gradual change procedures in behavior analysis,” Behav Analysis Practice, vol. 12, pp. 38–48, 03 2022.
  3. H. McCarty, D. Roth, K. Goode, J. Owen, L. Harrell, K. Donovan, and W. Haley, “Longitudinal course of behavioral problems during alzheimer’s disease: Linear versus curvilinear patterns of decline,” The journals of gerontology. Series A, Biological sciences and medical sciences, vol. 55, pp. M200–6, 05 2000.
  4. K. A. Cradock, L. R. Quinlan, F. M. Finucane, H. Gainforth, . Martin Ginis, Kathleen A, E. B.-N. Sanders, and G. ÓLaighin, “Design of a planner-based intervention to facilitate diet behaviour change in type 2 diabetes,” Apr 2022.
  5. G. Sprint, D. Cook, R. Fritz, and M. Schmitter-Edgecombe, “Detecting health and behavior change by analyzing smart home sensor data,” in 2016 IEEE Int. Conf. on Smart Comp., 2016, pp. 1–3.
  6. G. Civitarese, “Behavioral monitoring in smart-home environments for health-care applications,” in IEEE Int. Conf. on Pervasive Comp. and Comm. Workshops, 2017, pp. 105–106.
  7. A. A. Cook, G. Misirli, and Z. Fan, “Anomaly detection for iot time-series data: A survey,” IEEE Internet of Things Journal, vol. 7, pp. 6481–6494, 2020.
  8. V. Chandola, A. Banerjee, and V. Kumar, “Anomaly detection: A survey,” ACM Computing Surveys (CSUR), vol. 41, no. 3, p. 15, 2009.
  9. A. Sujith, G. S. Sajja, V. Mahalakshmi, S. Nuhmani, and B. Prasanalakshmi, “Systematic review of smart health monitoring using deep learning and artificial intelligence,” Neuroscience Informatics, vol. 2, no. 3, 2022, multimedia-based Emerging Tech. and Data Analytics for Neuroscience as a Service (NaaS).
  10. T. S. Sethi and M. Kantardzic, “Don’t pay for validation: Detecting drifts from unlabeled data using margin density,” Procedia Computer Science, vol. 53, pp. 103–112, 2015.
  11. D. M. dos Reis, P. Flach, S. Matwin, and G. Batista, “Fast unsupervised online drift detection using incremental kolmogorov-smirnov test,” in Proc. of the 22nd ACM SIGKDD Int. Conf. on Knowl. Disc. and Data Mining, 2016, pp. 1545–1554.
  12. A. Haque, L. Khan, and M. Baron, “Sand: Semi-supervised adaptive novel class detection and classification over data stream,” in Proc. of the AAAI Conf. on Artificial Intelligence, vol. 30, no. 1, 2016.
  13. Y. Kim and C. H. Park, “An efficient concept drift detection method for streaming data under limited labeling,” IEICE Trans. on Info. and Sys., vol. 100, no. 10, pp. 2537–2546, 2017.
  14. Y. S. Koh, “Cd-tds: Change detection in transactional data streams for frequent pattern mining,” in 2016 Int. Joint Conf. on Neural Networks.   IEEE, 2016, pp. 1554–1561.
  15. E. Lughofer, E. Weigl, W. Heidl, C. Eitzinger, and T. Radauer, “Recognizing input space and target concept drifts in data streams with scarcely labeled and unlabelled instances,” Information Sciences, vol. 355, pp. 127–151, 2016.
  16. S. A. Bashir, A. Petrovski, and D. Doolan, “A framework for unsupervised change detection in activity recognition,” Int. J. of Pervasive Comp. and Comm., 2017.
  17. T. S. Sethi and M. Kantardzic, “On the reliable detection of concept drift from streaming unlabeled data,” Expert Systems with Applications, vol. 82, pp. 77–99, 2017.
  18. A. F. J. Costa, R. A. S. Albuquerque, and E. M. dos Santos, “A drift detection method based on active learning,” in Int. Joint Conf. on Neural Networks.   IEEE, 2018, pp. 1–8.
  19. A. Liu, J. Lu, F. Liu, and G. Zhang, “Accumulating regional density dissimilarity for concept drift detection in data streams,” Pattern Recognition, vol. 76, pp. 256–272, 2018.
  20. T. S. Sethi and M. Kantardzic, “Handling adversarial concept drift in streaming data,” Expert systems with applications, vol. 97, pp. 18–40, 2018.
  21. W. Chang, C. Li, Y. Yang, and B. Póczos, “Kernel change-point detection with auxiliary deep generative models,” in 7th Int. Conf. on Learning Representations.   OpenReview.net, 2019.
  22. Ö. Gözüaçık, A. Büyükçakır, H. Bonab, and F. Can, “Unsupervised concept drift detection with a discriminative classifier,” in 28th ACM Int. Conf. on Info. and Knowl. Management, 2019, pp. 2365–2368.
  23. B. Li, Y.-j. Wang, D.-s. Yang, Y.-m. Li, and X.-k. Ma, “Faad: an unsupervised fast and accurate anomaly detection method for a multi-dimensional sequence over data stream,” Frontiers of Info. Tech. & Electronic Eng., vol. 20, no. 3, pp. 388–404, 2019.
  24. R. F. de Mello, Y. Vaz, C. H. Grossi, and A. Bifet, “On learning guarantees to unsupervised concept drift detection on data streams,” Expert Systems with Applications, vol. 117, pp. 90–102, 2019.
  25. F. Pinagé, E. M. dos Santos, and J. Gama, “A drift detection method based on dynamic classifier selection,” Data Mining and Knowl. Disc., vol. 34, no. 1, pp. 50–74, 2020.
  26. J. Haug and G. Kasneci, “Learning parameter distributions to detect concept drift in data streams,” in 25th Int. Conf. on Pattern Recognition.   IEEE, 2021, pp. 9452–9459.
  27. V. Cerqueira, H. M. Gomes, A. Bifet, and L. Torgo, “Studd: a student–teacher method for unsupervised concept drift detection,” Machine Learning, pp. 1–28, 2022.
  28. J. Haug, A. Braun, S. Zürn, and G. Kasneci, “Change detection for local explainability in evolving data streams,” in 31st ACM Int. Conf. on Information & Knowl. Management, 2022, pp. 706–716.
  29. R. N. Gemaque, A. F. J. Costa, R. Giusti, and E. M. dos Santos, “An overview of unsupervised drift detection methods,” WIREs Data Mining Knowl. Discov., vol. 10, no. 6, 2020.
  30. J. Lu, A. Liu, F. Dong, F. Gu, J. Gama, and G. Zhang, “Learning under concept drift: A review,” IEEE Trans. on Knowl. and Data Eng., vol. 31, no. 12, pp. 2346–2363, 2018.
  31. N. A. Barbosa, L. Travé-Massuyès, and V. H. Grisales, “A novel algorithm for dynamic clustering: properties and performance,” in 15th IEEE Int. Conf. on Machine Learning and Appl., 2016, pp. 565–570.
  32. D. Aragona, L. Podo, B. Prenkaj, and P. Velardi, “CoRoNNa: a deep sequential framework to predict epidemic spread,” in Proc. of the 36th Annual ACM Symposium on Applied Computing, 2021, pp. 10–17.
  33. B. Prenkaj, D. Distante, S. Faralli, and P. Velardi, “Hidden space deep sequential risk prediction on student trajectories,” Future Generation Computer Systems, vol. 125, pp. 532–543, 2021.
  34. B. Prenkaj, D. Aragona, A. Flaborea, F. Galasso, S. Gravina, L. Podo, E. Reda, and P. Velardi, “A self-supervised algorithm to detect signs of social isolation in the elderly from daily activity sequences,” Artificial Intelligence in Medicine, vol. 135, p. 102454, 2023.
  35. N. B. Roa, L. Travé-Massuyès, and V. H. Grisales-Palacio, “Dyclee: Dynamic clustering for tracking evolving environments,” Pattern Recognition, vol. 94, pp. 162–186, 2019.
  36. N. Chawla, S. Eschrich, and L. O. Hall, “Creating ensembles of classifiers,” in Proc. of the IEEE Int. Conf. on Data Mining, 2001, pp. 580–581.
  37. H. Alemdar, H. Ertan, O. D. Incel, and C. Ersoy, “Aras human activity datasets in multiple homes with multiple residents,” in 7th Int. Conf. on Pervasive Comp. Tech. for Healthcare and Workshops, 2013, pp. 232–235.
  38. D. J. Cook, “Learning setting-generalized activity models for smart spaces,” IEEE Int. Sys., vol. 2010, no. 99, p. 1, 2010.
  39. E. M. Tapia, S. S. Intille, and K. Larson, “Activity recognition in the home using simple and ubiquitous sensors,” in Int. Conf. on Pervasive Comp., 2004, pp. 158–175.
  40. T. Van Kasteren, G. Englebienne, and B. J. Kröse, “Activity recognition using semi-markov models on real world smart home datasets,” J. of Ambient Intelligence and Smart Envs., vol. 2, no. 3, pp. 311–325, 2010.
  41. A. Masciadri, F. Veronese, S. Comai, I. Carlini, and F. Salice, “Disseminating synthetic smart home data for advanced applications.” in CIKM Workshops, 2018.
  42. L. Podo and P. Velardi, “Anomalybyclick: An interactive visualization tool for monitoring activities of daily living and anomaly annotation,” in Int. Conf. on Advanced Vis. Interfaces, 2022, pp. 1–3.
  43. P. Fryzlewicz, “Wild binary segmentation for multiple change-point detection,” The Annals of Statistics, vol. 42, no. 6, pp. 2243–2281, 2014.
  44. E. Keogh, S. Chu, D. Hart, and M. Pazzani, “An online algorithm for segmenting time series,” in Proc. IEEE Int. Conf. on Data Mining, 2001, pp. 289–296.
  45. R. Killick, P. Fearnhead, and I. A. Eckley, “Optimal detection of changepoints with a linear computational cost,” J. of the American Stat. Assoc., vol. 107, no. 500, pp. 1590–1598, 2012.
  46. G. D. Wambui, G. A. Waititu, and A. K. Wanjoya, “The power of the pruned exact linear time(pelt) test in multiple changepoint detection,” American J. of Theoretical and Applied Stat., vol. 4, p. 581, 2015.
  47. C. Truong, L. Oudre, and N. Vayatis, “Selective review of offline change point detection methods,” Signal Processing, vol. 167, p. 107299, 2020.
  48. S. Arlot, A. Celisse, and Z. Harchaoui, “A kernel multiple change-point algorithm via model selection,” Journal of machine learning research, vol. 20, no. 162, 2019.
  49. J. Bai, “Estimating multiple breaks one at a time,” Econometric theory, vol. 13, no. 3, pp. 315–352, 1997.
  50. M. Miller, “The role of sleep and sleep disorders in the development, diagnosis, and management of neurocognitive disorders,” Frontiers in Neurology, vol. 6, 11 2015.
  51. P. Domingos and G. Hulten, “Mining high-speed data streams,” in Proceedings of the sixth ACM SIGKDD international conference on Knowledge discovery and data mining, 2000, pp. 71–80.
  52. A. Karasmanoglou, M. Antonakakis, and M. Zervakis, “Ecg-based semi-supervised anomaly detection for early detection and monitoring of epileptic seizures,” International Journal of Environmental Research and Public Health, vol. 20, no. 6, p. 5000, 2023.
  53. C. Liddell, C. Morris, B. Gray, A. Czerwinska, and B. Thomas, “Excess winter mortality associated with alzheimer’s disease and related dementias in the uk: A case for energy justice,” Energy Research & Social Science, vol. 11, pp. 256–262, 2016. [Online]. Available: https://www.sciencedirect.com/science/article/pii/S2214629615300724
Citations (2)
View on Semantic Scholar

Summary

We haven't generated a summary for this paper yet.

Ai Sparkles Streamline Icon: https://streamlinehq.com Summarize Now