A MODIFIED K-NEAREST NEIGHBOR CLASSIFIER TO DEAL WITH UNBALANCED CLASSES

Akram AlSukker, Ahmed Al-Ani, Amir Atiya

2009

Abstract

We present in this paper a simple, yet valuable improvement to the traditional k-Nearest Neighbor (kNN) classifier. It aims at addressing the issue of unbalanced classes by maximizing the class-wise classification accuracy. The proposed classifier also gives the option of favoring a particular class through evaluating a small set of fuzzy rules. When tested on a number of UCI datasets, the proposed algorithm managed to achieve a uniformly good performance.

References

  1. Cover, T. & Hart, P. (1967) Nearest neighbor pattern classification. Information Theory, IEEE Transactions on, 13, 21-27.
  2. Denoeux, T. (1995) A k-nearest neighbor classification rule based on Dempster-Shafer theory. Systems, Man and Cybernetics, IEEE Transactions on, 25, 804-813.
  3. Duda, R. O. & Hart, P. E. (1973) Pattern classification and scene analysis, N.Y., Wiley.
  4. Dudani, S. A. (1976) The Distance-Weighted k-NearestNeighbor Rule. SMC, 6, 325-327.
  5. Newman, A. A. A. D. J. (2007) UCI Machine Learning Repository. University of California, Irvine, School of Information and Computer Sciences.
  6. Paredes, R. & Vidal, E. (2006) Learning weighted metrics to minimize nearest-neighbor classification error. Pattern Analysis and Machine Intelligence, IEEE Transactions on, 28, 1100- 1110.
  7. Tan, S. (2005) Neighbor-weighted K-nearest neighbor for unbalanced text corpus. Expert Systems with Applications, 28, 667-671.
  8. Westin, L. K. (2001) Receiver operating characteristic (ROC) analysis. Department of Computing Science,UmeƄ University,sweden.
  9. Yong, Z., Yupu, Y. & Liang, Z. (2009) Pseudo nearest neighbor rule for pattern classification. Expert Syst. Appl., 36, 3587-3595.
Download


Paper Citation


in Harvard Style

AlSukker A., Al-Ani A. and Atiya A. (2009). A MODIFIED K-NEAREST NEIGHBOR CLASSIFIER TO DEAL WITH UNBALANCED CLASSES . In Proceedings of the International Joint Conference on Computational Intelligence - Volume 1: ICNC, (IJCCI 2009) ISBN 978-989-674-014-6, pages 408-413. DOI: 10.5220/0002282604080413


in Bibtex Style

@conference{icnc09,
author={Akram AlSukker and Ahmed Al-Ani and Amir Atiya},
title={A MODIFIED K-NEAREST NEIGHBOR CLASSIFIER TO DEAL WITH UNBALANCED CLASSES},
booktitle={Proceedings of the International Joint Conference on Computational Intelligence - Volume 1: ICNC, (IJCCI 2009)},
year={2009},
pages={408-413},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0002282604080413},
isbn={978-989-674-014-6},
}


in EndNote Style

TY - CONF
JO - Proceedings of the International Joint Conference on Computational Intelligence - Volume 1: ICNC, (IJCCI 2009)
TI - A MODIFIED K-NEAREST NEIGHBOR CLASSIFIER TO DEAL WITH UNBALANCED CLASSES
SN - 978-989-674-014-6
AU - AlSukker A.
AU - Al-Ani A.
AU - Atiya A.
PY - 2009
SP - 408
EP - 413
DO - 10.5220/0002282604080413