Orthogonal Decision Trees for Resource-Constrained Physiological Data Stream Monitoring Using Mobile Devices
This paper considers the problem of monitoring physiological data streams obtained from resource-constrained wearable sensing devices for pervasive health-care management. It considers Orthogonal decision trees (ODTs) that offer an effective way to construct a redundancy-free, accurate, and meaningful representation of large decision-tree-ensembles often created by popular techniques such as Bagging, Boosting, Random Forests and many distributed and data stream mining algorithms. ODTs are functionally orthogonal to each other and they correspond to the principal components of the underlying function space. This paper offers experimental results to document the performance of ODTs on grounds of accuracy, model complexity, and resource consumption.
KeywordsDecision Tree Random Forest Data Stream West Nile Virus Fourier Spectrum
Unable to display preview. Download preview PDF.
- 2.Quinlan, J.R.: C4.5: Programs for Machine Learning. Morgan Kauffman, San Francisco (1993)Google Scholar
- 4.Drucker, H., Cortes, C.: Boosting decision trees. Advances in Neural Information Processing Systems 8, 479–485 (1996)Google Scholar
- 8.Street, W.N., Kim, Y.: A streaming ensemble algorithm (sea) for large-scale classificaiton. In: Seventh ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, San Francisco, CA (2001)Google Scholar
- 10.Kargupta, H., Dutta, H.: Orthogonal Decision Trees. In: Fourth IEEE International Conference on Data Mining (ICDM), pp. 427–430 (2004)Google Scholar
- 12.Park, B.H., Kargupta, H.: Constructing simpler decision trees from ensemble models using Fourier analysis. In: Proceedings of the 7th Workshop on Research Issues in Data Mining and Knowledge Discovery, ACM SIGMOD, pp. 18–23 (2002)Google Scholar