Eta Correlation Coefficient Based Feature Selection Algorithm for Machine Learning: E-Score Feature Selection Algorithm

Abstract

Feature selection algorithms are of great importance in the field of machine learning. Significant reduction of very large data is the main function of feature selection algorithms. These methods are still being developed today. The reason for this is that data structures are growing day by day. As the data increases, more advanced, better performance, feature selection algorithms are needed. In this study, Eta Correlation Coefficient based E-Score Feature selection algorithm was developed. Two versions were prepared for E-Score. We tested the performance of the E-Score method with three classifiers and compared with conventional F-Score Feature Selection Algorithm. According to the results, both versions of the E-Score feature selection algorithm have improved performance and is better than the F-Score. According to these results, it is thought that the E-Score Feature Selection Algorithm can be used in the field of machine learning.

Keywords

• Eta Correlation Coefficient,E-Score Feature Selection Algorithm,Feature Selection Methods

References

1. [1] D. Guan, W. Yuan, Y.-K. Lee, K. Najeebullah, and M. K. Rasel, “A Review of Ensemble Learning Based Feature Selection,” IETE Tech. Rev., vol. 31, no. 3, pp. 190–198, 2014.
2. [2] K. Polat and S. Güneş, “A new feature selection method on classification of medical datasets: Kernel F-score feature selection,” Expert Syst. Appl., vol. 36, no. 7, pp. 10367–10373, Sep. 2009.
3. [3] A. R. Kavsaoğlu, K. Polat, and M. R. Bozkurt, “A novel feature ranking algorithm for biometric recognition with PPG signals.,” Comput. Biol. Med., vol. 49, pp. 1–14, Jun. 2014.
4. [4] J. Cai, J. Luo, S. Wang, and S. Yang, “Feature selection in machine learning: A new perspective,” Neurocomputing, vol. 300, pp. 70–79, Jul. 2018.
5. [5] T. Khoshgoftaar, D. Dittman, R. Wald, and A. Fazelpour, “First Order Statistics Based Feature Selection: A Diverse and Powerful Family of Feature Seleciton Techniques,” in 2012 11th International Conference on Machine Learning and Applications, 2012, pp. 151–157.
6. [6] A. Goltsev and V. Gritsenko, “Investigation of efficient features for image recognition by neural networks,” Neural Networks, vol. 28, pp. 15–23, Apr. 2012.
7. [7] H. Elghazel and A. Aussem, “Unsupervised feature selection with ensemble learning,” Mach. Learn., vol. 98, no. 1–2, pp. 157–180, Jan. 2015.
8. [8] Y. Li, S.-Y. Gao, and S. Chen, “Ensemble Feature Weighting Based on Local Learning and Diversity,” AAAI, 2012.

9. [9] D.-S. Huang, “Radial Basis Probabilistic Neural Networks: Model and Application,” Int. J. Pattern Recognit. Artif. Intell., vol. 13, no. 07, pp. 1083–1101, Nov. 1999.
10. [10] Tsang-Hsiang Cheng, Chih-Ping Wei, and V. S. Tseng, “Feature Selection for Medical Data Mining: Comparisons of Expert Judgment and Automatic Approaches,” in 19th IEEE Symposium on Computer-Based Medical Systems (CBMS’06), 2006, pp. 165–170.
11. [11] E. C. Andrzejak RG, Lehnertz K, Rieke C, Mormann F, David P, “UCI Machine Learning Repository: Epileptic Seizure Recognition Data Set,” UCI, 2001. [Online]. Available: https://archive.ics.uci.edu/ml/datasets/Epileptic+Seizure+Recognition. [Accessed: 14-Aug-2018].
12. [12] R. G. Andrzejak, K. Lehnertz, F. Mormann, C. Rieke, P. David, and C. E. Elger, “Indications of nonlinear deterministic and finite-dimensional structures in time series of brain electrical activity: Dependence on recording region and brain state,” Phys. Rev. E - Stat. Physics, Plasmas, Fluids, Relat. Interdiscip. Top., vol. 64, no. 6, p. 8, Nov. 2001.
13. [13] R. Alpar, Applied Statistic and Validation - Reliability. Detay Publishing, 2010.
14. [14] R. Koch, The 80/20 Principle and 92 Other Powerful Laws of Nature: The Science of Success. 2014.