Ensemble Gene Selection Method Based on Multiple Tree Models
2
조회
0
다운로드
0
인용
0
참고문헌
키워드
초록
Identifying highly discriminating genes is a critical step in tumor recognition tasks based on microarray geneexpression profile data and machine learning. Gene selection based on tree models has been the subject ofseveral studies. However, these methods are based on a single-tree model, often not robust to ultra-highdimensionalmicroarray datasets, resulting in the loss of useful information and unsatisfactory classificationaccuracy. Motivated by the limitations of single-tree-based gene selection, in this study, ensemble gene selectionmethods based on multiple-tree models were studied to improve the classification performance of tumoridentification. Specifically, we selected the three most representative tree models: ID3, random forest, andgradient boosting decision tree. Each tree model selects top-n genes from the microarray dataset based on itsintrinsic mechanism. Subsequently, three ensemble gene selection methods were investigated, namely multipletreemodel intersection, multiple-tree module union, and multiple-tree module cross-union, were investigated. Experimental results on five benchmark public microarray gene expression datasets proved that the multipletree module union is significantly superior to gene selection based on a single tree model and other competitivegene selection methods in classification accuracy.
Abstract
Identifying highly discriminating genes is a critical step in tumor recognition tasks based on microarray geneexpression profile data and machine learning. Gene selection based on tree models has been the subject ofseveral studies. However, these methods are based on a single-tree model, often not robust to ultra-highdimensionalmicroarray datasets, resulting in the loss of useful information and unsatisfactory classificationaccuracy. Motivated by the limitations of single-tree-based gene selection, in this study, ensemble gene selectionmethods based on multiple-tree models were studied to improve the classification performance of tumoridentification. Specifically, we selected the three most representative tree models: ID3, random forest, andgradient boosting decision tree. Each tree model selects top-n genes from the microarray dataset based on itsintrinsic mechanism. Subsequently, three ensemble gene selection methods were investigated, namely multipletreemodel intersection, multiple-tree module union, and multiple-tree module cross-union, were investigated. Experimental results on five benchmark public microarray gene expression datasets proved that the multipletree module union is significantly superior to gene selection based on a single tree model and other competitivegene selection methods in classification accuracy.
참고문헌 (23)
- Haidi Rao, "Feature selection based on artificial bee colony and gradient boosting decision tree", Applied Soft Computing, 74, pp.634~642 (2019), Elsevier BV[DOI]
- Jorng-Tzong Horng, "An expert system to classify microarray gene expression data using gene selection by decision tree", Expert Systems with Applications, 36(5), pp.9072~9081 (2009), Elsevier BV[DOI]
- W, Xiong, "A hybrid improved ant colony optimization and random forests feature selection method for microarray data", Proceedings of 2009 5th International Joint Conference on INC, IMS and IDC, pp.559~563 (2019)
- Guesh Dagnew, "Ensemble learning‐based classification of microarray cancer data on tree‐based features", Cognitive Computation and Systems, 3(1), pp.48~60 (2021), Institution of Engineering and Technology (IET)[DOI]
- Xiongshi Deng, "Hybrid gene selection approach using XGBoost and multi-objective genetic algorithm for cancer classification", Medical & Biological Engineering & Computing, 60(3), pp.663~681 (2022), Springer Science and Business Media LLC[DOI]
- T. Chen, "XGBoost: a scalable tree boosting system", Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, pp.785~794 (2016)
- J.R. Quinlan, "Induction of decision trees", Machine Learning, 1(1), pp.81~106 (1986), Springer Science and Business Media LLC[DOI]
- Leo Breiman, "Random forests", Machine Learning, 45(1), pp.5~32 (2001), Springer Science and Business Media LLC[DOI]
- Jerome H. Friedman, "Stochastic gradient boosting", Computational Statistics & Data Analysis, 38(4), pp.367~378 (2002), Elsevier BV[DOI]
- Chih-Chung Chang, "LIBSVM: a library for support vector machines", ACM Transactions on Intelligent Systems and Technology, 2(3), pp.1~27 (2011), Association for Computing Machinery (ACM)[DOI]
- G. H. John, "Estimating continuous distributions in Bayesian classifiers", Proceedings of the 11th Annual Conference on Uncertainty in Artificial Intelligence, pp.338~345 (1995)
- J. R. Quinlan, "C4.5: Programs for Machine Learning" (2014), Morgan Kaufmann Publishers
- A. D. Gordon, "Classification and Regression Trees.", Biometrics, 40(3), pp.874~874 (1984), JSTOR[DOI]
- J. H. Friedman, "Greedy function approximation: a gradient boosting machine", Annals of Statistics, 29(5), pp.1189~1232 (2001)
- F. Borovecki, "Genome-wide expression profiling of human blood reveals biomarkers for Huntington's disease", Proceedings of the National Academy of Sciences, 102(31), pp.11023~11028 (2005), Proceedings of the National Academy of Sciences[DOI]
- Ash A. Alizadeh, "Distinct types of diffuse large B-cell lymphoma identified by gene expression profiling", Nature, 403(6769), pp.503~511 (2000), Springer Science and Business Media LLC[DOI]
- Tao Li, "A comparative study of feature selection and multiclass classification methods for tissue classification based on gene expression", Bioinformatics, 20(15), pp.2429~2437 (2004), Oxford University Press (OUP)[DOI]
- Arindam Bhattacharjee, "Classification of human lung carcinomas by mRNA expression profiling reveals distinct adenocarcinoma subclasses", Proceedings of the National Academy of Sciences, 98(24), pp.13790~13795 (2001), Proceedings of the National Academy of Sciences[DOI]
- Zexuan Zhu, "Markov blanket-embedded genetic algorithm for gene selection", Pattern Recognition, 40(11), pp.3236~3248 (2007), Elsevier BV[DOI]
- Qasem Al-Tashi, "Binary Optimization Using Hybrid Grey Wolf Optimization for Feature Selection", IEEE Access, 7, pp.39496~39508 (2019), Institute of Electrical and Electronics Engineers (IEEE)[DOI]
- Xiaojuan Huang, "Feature clustering based support vector machine recursive feature elimination for gene selection", Applied Intelligence, 48(3), pp.594~607 (2017), Springer Science and Business Media LLC[DOI]
- Z. Hou, "A kernel discriminant information approach to non-linear feature selection", Proceedings of 2019 International Joint Conference on Neural Networks (IJCNN), pp.1~10 (2019)
- Ryan J. Urbanowicz, "Benchmarking relief-based feature selection methods for bioinformatics data mining", Journal of Biomedical Informatics, 85, pp.168~188 (2018), Elsevier BV[DOI]