In recent years, portable computing devices (PCDs) such as smart phones and tablet terminals have been popularized at a tremendous speed. People around the world are now using PCDs for different purposes. To resolve “digital divide” problem, it is desired to embed awareness agents (A-agents) that can recognize different situations, detect important information, and help human users make decisions efficiently and effectively. To use A-agents in one PCD, it is necessary to implement each agent with a reasonable cost. For this purpose, we can use dimensionality reduction (DR). To reduce the total cost, sophisticated DR methods cannot be used. In this paper, we investigate the performance changes of SVM-based A-agents, before and after centroid based DR. Experimental results show that in most cases the performance can be preserved with the properly chosen learning parameters.

1. [1] Q. F. Zhao, “An Introduction to Computational Awareness,” Proc. of Int. Conf. on Awareness Science and Technology (iCAST2012), Seoul, pp. 29-34, 2012.
2. [2] V. Vapnik, “Statistical Learning Theory,” New York, Wiley, 1998.
3. [3] T. Joachims, “Text categorization with support vector machines: Learning with many relevant features,” Proc. of ECML-98, 10th European Conf. on Machine Learning, No.1398, pp. 137-142, 1998.
4. [4] I. T. Jolliffe, “Principal Component Analysis,” Springer, 2nd edition, 2002.
5. [5] P. E. Hart, R. O. Duda, and D. G. Stork, “Pattern Classification,” Wiley-Interscience, 2nd edition, 2000.
6. [6] H. Hayashi and Q. F. Zhao, “Inducing compact NNTrees through discriminant multiple centroids based dimensionality reduction,” Int. J. of Innovative computing, Information and Control, Vol.7, No.5, 2011.
7. [7] C. L. Blake and C. J. Merz, “UCI Repository of Machine Learning Databases.”
   http://www.ics.uci.edu/˜mlearn/MLRepository.html [Accessed June 17, 2013]
8. [8] T. Joachims, “SVM-Light Support Vector Machine.”
   http://svmlight.joachims.org/ [Accessed August 14, 2008]