Integration of discriminative and generative models for activity recognition in smart homes

Article English OPEN
Fahad, L. G. ; Rajarajan, M. (2015)

Activity recognition in smart homes enables the remote monitoring of elderly and patients. In healthcare systems, reliability of a recognition model is of high importance. Limited amount of training data and imbalanced number of activity instances result in over-fitting thus making recognition models inconsistent. In this paper, we propose an activity recognition approach that integrates the distance minimization (DM) and probability estimation (PE) approaches to improve the reliability of recognitions. DM uses distances of instances from the mean representation of each activity class for label assignment. DM is useful in avoiding decision biasing towards the activity class with majority instances; however, DM can result in over-fitting. PE on the other hand has good generalization abilities. PE measures the probability of correct assignments from the obtained distances, while it requires a large amount of data for training. We apply data oversampling to improve the representation of classes with less number of instances. Support vector machine (SVM) is applied to combine the outputs of both DM and PE, since SVM performs better with imbalanced data and further improves the generalization ability of the approach. The proposed approach is evaluated using five publicly available smart home datasets. The results demonstrate better performance of the proposed approach compared to the state-of-the-art activity recognition approaches.
  • References (51)
    51 references, page 1 of 6

    [1] S. Mahmoud, A. Lotfi, C. Langensiepen, Behavioural pattern identification and prediction in intelligent environments, Applied Soft Computing 13 (4) (2013) 1813-1822.

    [2] L. Chen, C. Nugent, H. Wang, A knowledge-driven approach to activity recognition in smart homes, IEEE Trans. on Knowledge and Data Engineering 24 (6) (2012) 961-974.

    [3] P. Rashidi, D. J. Cook, L. B. Holder, M. Schmitter-Edgecombe, Discovering activities to recognize and track in a smart environment, IEEE Trans. on Knowledge and Data Engineering 23 (4) (2011) 527-539.

    [4] B. Chikhaoui, S. Wang, H. Pigot, Activity discovery and recognition by combining sequential patterns and latent dirichlet allocation, Pervasive and Mobile Computing 8 (6) (2012) 845-862.

    [5] L. G. Fahad, A. Khan, M. Rajarajan, Activity recognition in smart homes with self verification of assignments, Neurocomputing 149 (0) (2015) 1286-1298.

    [6] B. Das, C. Chen, N. Dasgupta, D. J. Cook, A. M. Seelye, Automated prompting in a smart home environment, in: Proc. of IEEE Int. Workshops on Data Mining, Sydney, Australia, 2010, pp. 1045-1052.

    [7] L. G. Fahad, S. F. Tahir, M. Rajarajan, Activity recognition in smart homes using clustering based classification, in: Proc. of IEEE Int. Conf. on Pattern Recognition, Stockholm, Sweden, 2014, pp. 1348-1353.

    [8] L. G. Fahad, A. Ali, M. Rajarajan, Long term analysis of daily activities in smart home, in: Proc. of the European Symp. on Artificial Neural Networks, Computational Intelligence and Machine Learning, Bruges, Belgium, 2013, pp. 419-424.

    [9] A. Fleury, M. Vacher, N. Noury, Svm-based multimodal classification of activities of daily living in health smart homes: sensors, algorithms, and first experimental results, IEEE Trans. on Information Technology in Biomedicine 14 (2) (2010) 274-283.

    [10] T. V. Kasteren, A. Noulas, G. Englebienne, B. Krose, Accurate activity recognition in a home setting, in: Proc. of Int. conf. on Ubiquitous computing, Seoul, Korea, 2008, pp. 1-9.

  • Metrics
    No metrics available
Share - Bookmark