Education, Science, Technology, Innovation and Life
Open Access
Sign In

A Novel Classification Model based on Ensemble Feature Selection and Hyper-parameter Optimation

Download as PDF

DOI: 10.23977/jaip.2022.050103 | Downloads: 13 | Views: 866

Author(s)

Liguo Zhang 1, Jian Zhang 1, Yan Wang 2

Affiliation(s)

1 College of Information Science & Technology, Agricultural University of Hebei, Hebei, Baoding, China
2 Dept of Computer, North China Electric Power University, Hebei, Baoding, China

Corresponding Author

Liguo Zhang

ABSTRACT

Background and purpose: the central problem in machine learning is identifying a representative set of features from which to construct a classification model and setting the best Hyper-Parameters of the model for a particular task. Materials and Methods: Feature selection aims to reduce the dimensionality of patterns for classificatory analysis by selecting the most informative instead of irrelevant and/or redundant features. Here, RFE is used for selecting the most useful predictive features, PCA extract the characteristics of all the original variables. And also the Grid Search is used for model HPO. Experiments are applied on Breast cancer dataset. Findings and Conclusion: The experiments show that, the number of features is reduced from 30 to 15 features and the classification accuracy increase from 95.6% to 97.02%. And the proposed Feature selection and hyper-parameter optimation method can be used to other pattern classification problems.

KEYWORDS

Feature Selection, PCA, HPO, Recursive Feature Elimination, Random Forest Classification

CITE THIS PAPER

Liguo Zhang, Jian Zhang and Yan Wang, A Novel Classification Model based on Ensemble Feature Selection and Hyper-parameter Optimation. Journal of Artificial Intelligence Practice (2022) Vol. 5: 11-17. DOI: http://dx.doi.org/10.23977/jaip.2022.050103.

REFERENCES

[1] Eid, H. F., Hassanien, A. E., Kim, T. H., & Banerjee, S. (2013). Linear Correlation-Based Feature Selection for Network Intrusion Detection Model. Springer Berlin Heidelberg. Springer Berlin Heidelberg.
[2] Kumaravel., V. ,  Raja., K. ,  Kumaravel., V. , &  Raja., K. . (2014). Feature Subset Selection Algorithm for High-Dimensional Data by using FAST Clustering Approach.
[3] Akadi, A. E., Ouardighi, A. E., & D Aboutajdine. (2008). A powerful feature selection approach based on mutual information. International journal of computer science & network security.
[4] Hall, M. A. (2000). Correlation-based feature selection for machine learning. Phd Thesis Waikato Univer Sity.
[5 Chaves, R., J Ramírez, JM Górriz, M López, & Segovia, F. (2009). Svm-based computer-aided diagnosis of the alzheimer's disease using t-test nmse feature selection with feature correlation weighting. Neuroscience Letters, 461(3), 293-297.
[6] Upadhyay, D., Manero, J., Zaman, M., & Sampalli, S. (2021). Intrusion detection in scada based power grids: recursive feature elimination model with majority vote ensemble algorithm. IEEE Transactions on Network Science and Engineering, PP (99), 1-1.
[7] Kim, J.  "A Comparative Study of Sequential Feature Selection Methods for Support Vector Machine." (2008).
[8] Naing, T. T., & Khaing, K. T. (2010). Enhanced features ranking and selection using recursive feature elimination(rfe) and k-nearest neighbor algorithms in support vector machine for intrusion detection system. International journal of computer science issues.
[9] Chen, X. W., and J. C. Jeong. "Enhanced recursive feature elimination." Machine Learning and Applications, 2007. ICMLA 2007. Sixth International Conference on IEEE, 2008.
[10] Zhang, L., Zhi, L., & Yi, L. I. (2013). Research on diagnosis method of breast tumor based on factor analysis and ga-bp. Journal of Convergence Information Technology.
[11] Karabatak, M., & Ince, M. C. (2009). An expert system for detection of breast cancer based on association rules and neural network. Expert Systems with Applications, 36(2p2), 3465-3469.

Downloads: 5766
Visits: 178119

Sponsors, Associates, and Links


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

Copyright © 2016 - 2031 Clausius Scientific Press Inc. All Rights Reserved.