Feature Dimension Reduction Using Euclidean Distance Oriented Similarity Based Rough Set Model

Authors

  • Mondal AC Department of Computer Science, University of Burdwan, Burdwan, India
  • Kolay S Electrical & Automation, SMS India Pvt. Ltd., Kolkata, India

Keywords:

Rough Set, Feature Dimension Reduction, Similarity Relation, KNN, Grammatical Facial Expression Recognition

Abstract

In machine learning, a very high dimensional data reduces the performance of a classifier. To overcome this, a relevant feature dimension reduction algorithm can be applied before applying any classification algorithm. Rough set theory [1] is a very good tool to reduce the feature dimension of an information system or decision system. However, if a decision system contains real-valued data, we cannot apply directly the rough set theory. Various extensions to rough set can be used to handle this kind of data. Among them, Fuzzy-Rough set theory [2], similarity based rough set model [3] are interesting. We propose an algorithm for dimension reduction using Euclidean distance oriented similarity based rough set model. To show the effectiveness of the algorithm, we take Grammatical Facial Expression Dataset from UCI Machine Learning Repository, created by Freitas et al. [4] and applied KNN classifier before and after feature dimension reduction.

References

[1] Z. Pawlak, “Rough Sets. Theoretical Aspects of Reasoning about Data”, Kluwer Academic Publishers, 1991.

[2] R. Jensen, Q. Shen, “Semantics-preserving dimensionality reduction: rough and fuzzy-rough-based approaches”, IEEE Transactions on Knowledge and Data Engineering, Vol.16, Issue.12, pp.1457-1471, 2004.

[3] J. Stepaniuk, “Similarity Based Rough Sets and Learning”, Proceedings of the Fourth International Workshop on Rough Sets, Fuzzy Sets, and Machine Discovery, Tokyo, Japan, pp. 18-22, 1996.

[4] F. A. Freitas, S. M.. Peres, C. A. M. Lima, F. V. Barbosa, “Grammatical Facial Expressions Recognition with Machine Learning”, 27th Florida Artificial Intelligence Research Society Conference (FLAIRS), Palo Alto, pp.180-185, 2014.

[5] K. S. Ray, S. Kolay “Application of Approximate Equality for Reduction of Feature Vector Dimension”, Journal of Pattern Recognition Research, Vol.11, Issue. 1, pp.26-40, 2016

Downloads

Published

2025-11-24

How to Cite

[1]
A. Mondal and S. Kolay, “Feature Dimension Reduction Using Euclidean Distance Oriented Similarity Based Rough Set Model”, Int. J. Comp. Sci. Eng., vol. 7, no. 1, pp. 33–36, Nov. 2025.

Issue

Vol. 7 No. 1 (2019): IJCSE Special Issue Jan Edition

Section

Research Article

License

Creative Commons License

This work is licensed under a Creative Commons Attribution 4.0 International License.

Authors contributing to this journal agree to publish their articles under the Creative Commons Attribution 4.0 International License, allowing third parties to share their work (copy, distribute, transmit) and to adapt it, under the condition that the authors are given credit and that in the event of reuse or distribution, the terms of this license are made clear.