Removal of Inconsistent Training Data in Electronic Nose Using Rough Set
Inconsistency in the electronic nose data set may appear due to noise that originates from various sources like electrical equipments, measuring instruments and some times the process itself. The presence of high noise leads to produce data that are of conflicting decision and thus encounters misleading or biased results. Also the performance of the electronic nose depends upon the number of relevant, irredundant features present in the data set. In an electronic nose the features correspond to the sensor array. While deploying an electronic nose for a specific application, it is observed that some of the features (sensors response) may not be required rather than only a subset of the sensor array contributes to the decision, which implies the optimization of sensor array is also important. To obtain a consistent precise data set both the conflicting data and irrelevant features must be removed. The rough set theory proposed by Z. Pawlak, is capable of dealing with such an imprecise, inconsistent data set and in this paper, the rough-set based algorithm has been applied to remove the conflicting training patterns and optimize the sensor array in an electronic nose instrument used for sensing aroma of black tea samples.
KeywordsElectronic nose feature selection reduct rough set sensor array
Unable to display preview. Download preview PDF.
- 1.Natale, C.D., Macagnano, A., Mantini, A., Davide, F., D’Amnico, A., Paolesse, R., Boschi, T., Faccio, M., Ferri, G.: Advances in food analysis by electronic nose. In: Proc. of the IEEE Int. Symp. on Industrial Electronics, vol. 1, pp. 122–127 (1997)Google Scholar
- 2.Pawlak, Z.: Rough set theory and its applications. J. Telecom. Inform. Techno. 3, 7–10 (2002)Google Scholar
- 8.Yang, P., Li, J., Huang, Y.: An attribute reduction algorithm by rough set based on binary discernibility matrix. In: Proc. of the Fifth In. Conf. Fuzzy Systems and Knowledge Discovery, vol. 2, pp. 276–280 (October 2000)Google Scholar
- 9.Nguyen, S.H., Skowron, A.: Quantization of real value attributes, Rough set and boolean reasoning approach. In: Proc. of the Second Joint Annual Conf. on Information Science, Wrightsville Beach, North Carolina, pp. 34–37 (1995)Google Scholar
- 10.Dai, J.-H., Li, Y.-X.: Study on discretization based on rough set theory. In: Proc. of the First Int. Conf. on Machine Learning and Cybernetics, Beijing, pp. 1371–1373 (November 2002)Google Scholar
- 14.Duda, R.O., Stork, D.G., Hart, P.E.: Pattern classification, 2nd edn., p. 115. John Wiley and Sons (2001)Google Scholar
- 15.Haykin, S.: Neural Networks: A Comprehensive Foundation, 2nd edn. Pearson Education, Asia (2001)Google Scholar