Abstract
In machine learning, randomness is a crucial factor in the success of ensemble learning, and it can be injected into tree-based ensembles by rotating the feature space. However, it is a common practice to rotate the feature space randomly. Thus, a large number of trees are required to ensure the performance of the ensemble model. This random rotation method is theoretically feasible, but it requires massive computing resources, potentially restricting its applications. A multimodal genetic algorithm based rotation forest (MGARF) algorithm is proposed in this paper to solve this problem. It is a tree-based ensemble learning algorithm for classification, taking advantage of the characteristic of trees to inject randomness by feature rotation. However, this algorithm attempts to select a subset of more diverse and accurate base learners using the multimodal optimization method. The classification accuracy of the proposed MGARF algorithm was evaluated by comparing it with the original random forest and random rotation ensemble methods on 23 UCI classification datasets. Experimental results show that the MGARF method outperforms the other methods, and the number of base learners in MGARF models is much fewer.
摘要
在机器学习中, 随机性是集成学习成功的一个关键因素, 可通过旋转特征空间将随机性注入到基于树的集成学习模型中。随机地旋转特征空间是一种常见的做法, 因此, 需要大量的树来保证集成模型的性能。该方法在理论上是可行的, 但它需要大量的计算资源, 因此, 本文提出了基于多峰遗传算法的旋转森林(MGARF)算法。该方法是一种基于树的分类集成学习算法, 利用树的特性通过特征旋转注入随机性, 利用多峰优化方法选择一个更多样化、更准确的基学习机子集。通过在23 个UCI 分类数据集上与原始随机森林和随机旋转集成方法进行比较, 评估了所提出的MGARF 算法的分类精度。实验结果表明, MGARF 方法的性能优于其他方法, 而且MGARF 模型中的基学习机的数量更少。
Similar content being viewed by others
References
SAGI O, ROKACH L. Ensemble learning: A survey [J]. WIREs Data Mining and Knowledge Discovery, 2018, 8(4): e1249. DOI: https://doi.org/10.1002/widm.1249.
DIETTERICH T G. Machine-learning research: Four current directions [J]. AI Magazine, 1997, 18(4): 97–136. DOI: https://doi.org/10.1609/aimag.v18i4.1324.
DONG Xi-bin, YU Zhi-wen, CAO Wen-ming, SHI Yi-fan, MA Qian-li. A survey on ensemble learning [J]. Frontiers of Computer Science, 2020, 14(2): 241–258. DOI: https://doi.org/10.1007/s11704-019-8208-z.
YU Wan-ke, ZHAO Chun-hui. Online fault diagnosis for industrial processes with bayesian network-based probabilistic ensemble learning strategy [J]. IEEE Transactions on Automation Science and Engineering, 2019, 16(4): 1922–1932. DOI: https://doi.org/10.1109/TASE.2019.2915286.
WANG Zhen-ya, LU Chen, ZHOU Bo. Fault diagnosis for rotary machinery with selective ensemble neural networks [J]. Mechanical Systems and Signal Processing, 2018, 113(SI): 112–130. DOI: https://doi.org/10.1016/j.ymssp.2017.03.051.
XIA Chong-kun, SU Cheng-li, CAO Jiang-tao, LI Ping. MultiBoost with ENN-based ensemble fault diagnosis method and its application in complicated chemical process [J]. Journal of Central South University, 2016, 23(5): 1183–1197. DOI: https://doi.org/10.1007/s11771-016-0368-5.
SONG Yan, ZHANG Shu-jing, HE Bo, SHA Qi-xin, SHEN Yue, YAN Tian-hong, NIAN Rui, LENDASSE A. Gaussian derivative models and ensemble extreme learning machine for texture image classification [J]. Neurocomputing, 2018, 277(SI): 53–64. DOI: https://doi.org/10.1016/j.neucom.2017.01.113.
MAO Ke-ming, DENG Zhuo-fu. Lung nodule image classification based on ensemble machine learning [J]. Journal of Medical Imaging and Health Informatics, 2016, 6(7): 1679–1685. DOI: https://doi.org/10.1166/jmihi.2016.1871.
CHEN Cun-jian, DANTCHEVA A, ROSS A. An ensemble of patch-based subspaces for makeup-robust face recognition [J]. Information Fusion, 2016, 32: 80–92. DOI: https://doi.org/10.1016/j.inffus.2015.09.005.
YAMAN M A, SUBASI A, RATTAY F. Comparison of random subspace and voting ensemble machine learning methods for face recognition [J]. Symmetry, 2018, 10(11): 651. DOI: https://doi.org/10.3390/sym10110651.
KRAWCZYK B, MINKU L L, GAMA J, STEFANOWSKI J, WOZNIAK M. Ensemble learning for data stream analysis: A survey [J]. Information Fusion, 2017, 37: 132–156. DOI: https://doi.org/10.1016/j.inffus.2017.02.004.
PIETRUCZUK L, RUTKOWSKI L, JAWORSKI M, DUDA P. How to adjust an ensemble size in stream data mining? [J]. Information Sciences, 2017, 381: 46–54. DOI: https://doi.org/10.1016/j.ins.2016.10.028.
TUMER K, GHOSH J. Analysis of decision boundaries in linearly combined neural classifiers [J]. Pattern Recognition, 1996, 29(2): 341–348. DOI: https://doi.org/10.1016/0031-3203(95)00085-2.
BROWN G, WYATT J, HARRIS R, YAO Xin. Diversity creation methods: A survey and categorization [J]. Information Fusion, 2005, 6(1): 5–20. DOI: https://doi.org/10.1016/j.inffus.2004.04.004.
NGOC P V, NGOC C V T, NGOC T V T, DUY D N. A C4.5 algorithm for english emotional classification [J]. Evolving Systems, 2019, 10(3): 425–451. DOI: https://doi.org/10.1007/s12530-017-9180-1.
RAHBARI D, NICKRAY M. Task offloading in mobile fog computing by classification and regression tree [J]. Peer-to-peer Networking and Applications, 2020, 13(1): 104–122. DOI: https://doi.org/10.1007/s12083-019-00721-7.
RAO Hai-di, SHI Xian-zhang, RODRIGUE A K, FENG Juanjuan, XIA Ying-chun, ELHOSENY M, YUAN Xiao-hui, GU Li-chuan. Feature selection based on artificial bee colony and gradient boosting decision tree[J]. Applied Soft Computing, 2019, 74: 634–642. DOI: https://doi.org/10.1016/j.asoc.2018.10.036.
LI Mu-jin, XU Hong-hui, DENG Yong. Evidential decision tree based on belief entropy [J]. Entropy, 2019, 21(9). DOI:https://doi.org/10.3390/e21090897.
BREIMAN L. Random forests [J]. Machine Learning, 2001, 45(1): 5–32. DOI: https://doi.org/10.1023/A:1010933404324.
SCHONLAU M, ZOU R Y. The random forest algorithm for statistical learning [J]. The Stata Journal, 2020, 20(1): 3–29. DOI: https://doi.org/10.1177/1536867X20909688.
RODRÍGUEZ J J, KUNCHEVA L I, ALONSO C J. Rotation forest: A new classifier ensemble method [J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2006, 28(10): 1619–1630. DOI: https://doi.org/10.1109/TPAMI.2006.211.
BLASER R, FRYZLEWICZ P. Random rotation ensembles [J]. Journal of Machine Learning Research, 2016, 17(1): 1–26. [2020-01-13] https://www.jmlr.org/papers/volume17/blaser16a/blaser16a.pdf, 2021-1-12/.
PHAM H, OLAFSSON S. Bagged ensembles with tunable parameters [J]. Computational Intelligence, 2019, 35(1): 184–203. DOI: https://doi.org/10.1111/coin.12198.
HANSEN L K, SALAMON P. Neural network ensembles[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 1990, 12(10): 993–1001. DOI: https://doi.org/10.1109/34.58871.
TAMA B A, RHEE K H. Tree-based classifier ensembles for early detection method of diabetes: an exploratory study [J]. Artificial Intelligence Review, 2019, 51(3): 355–370. DOI: https://doi.org/10.1007/s10462-017-9565-3.
ZHOU Zhi-hua, WU Jian-xin, TANG Wei. Ensembling neural networks: Many could be better than all [J]. Artificial Intelligence, 2002, 137(1, 2): 239–263. DOI: https://doi.org/10.1016/S0004-3702(02)00190-X.
ZENG Xiao-dong, WONG D F, CHAO L S. Constructing better classifier ensemble based on weighted accuracy and diversity measure [J]. The Scientific World Journal, 2014: 961747. DOI: https://doi.org/10.1155/2014/961747.
PEIMANKAR A, WEDDELL S J, JALAL T, LAPTHORN A C. Multi-objective ensemble forecasting with an application to power transformers [J]. Applied Soft Computing, 2018, 68: 233–248. DOI: https://doi.org/10.1016/j.asoc.2018.03.042.
PETROWSKI A. A clearing procedure as a niching method for genetic algorithms [C]// Proceedings of IEEE International Conference on Evolutionary Computation. Nagoya Japan: IEEE, 1996: 798–803. DOI: https://doi.org/10.1109/ICEC.1996.542703.
MENGSHOEL O J, GOLDBERG D E. The crowding approach to niching in genetic algorithms [J]. Evolutionary Computation, 2008, 16(3): 315–354. DOI: https://doi.org/10.1162/evco.2008.16.3.315.
GOLDBERG D E, RICHARDSON J. Genetic algorithms with sharing for multimodal function optimization [C]// Proceedings of the Second International Conference on Genetic Algorithms on Genetic Algorithms and Their Application. Cambridge, MA, USA: L. Erlbaum Associates, 1987: 41–49. [2021-01-13] https://dl.acm.org/doi/10.5555/42512.42519, 2021-1-13/.
LI Jian-ping, BALAZS M E, PARKS G T, CLARKSON P J. A species conserving genetic algorithm for multimodal function optimization [J]. Evolutionary Computation, 2002, 10(3): 207–234. DOI: https://doi.org/10.1162/106365602760234081.
THOMSEN R. Multimodal optimization using crowding-based differential evolution [C]// Proceedings of the 2004 Congress on Evolutionary Computation. Portland, USA: IEEE, 2004(2): 1382–1389. DOI: https://doi.org/10.1109/CEC.2004.1331058.
LI Xiao-dong. Efficient differential evolution using speciation for multimodal function optimization [C]// Proceedings of the 7th Annual Conference on Genetic and Evolutionary Computation. New York, USA: Association for Computing Machinery, 2005: 873–880. DOI: https://doi.org/10.1145/1068009.1068156.
LI Wei, FAN Yao-chi, XU Qing-zheng. Evolutionary multimodal optimization based on bi-population and multimutation differential evolution [J]. International Journal of Computational Intelligence Systems, 2020, 13(1):1345–1367. DOI: https://doi.org/10.2991/ijcis.d.200826.001.
WANG Zi-jia, ZHAN Zhi-hui, LIN Ying, YU Wei-jie, WANG Hua, KWONG S, ZHANG Jun. Automatic niching differential evolution with contour prediction approach for multimodal optimization problems [J]. IEEE Transactions on Evolutionary Computation, 2020, 24(1): 114–128. DOI: https://doi.org/10.1109/TEVC.2019.2910721.
LIU Qing-xue, DU Sheng-zhi, van WYK B J, SUN Yan-xia. Niching particle swarm optimization based on Euclidean distance and hierarchical clustering for multimodal optimization [J]. Nonlinear Dynamics, 2020, 99(3): 2459–2477. DOI: https://doi.org/10.1007/s11071-019-05414-7.
WANG Zi-jia, ZHAN Zhi-hui, LIN Ying, YU Wei-jie, YUAN Hua-qiang, GU Tian-long, KWONG S, ZHANG Jun. Dual-strategy differential evolution with affinity propagation clustering for multimodal optimization problems [J]. IEEE Transactions on Evolutionary Computation, 2018, 22(6): 894–908. DOI: https://doi.org/10.1109/TEVC.2017.2769108.
LIU Qing-xue, DU Sheng-zhi, van WYK B J, SUN Yan-xia. Double-layer-clustering differential evolution multimodal optimization by speciation and self-adaptive strategies [J]. Information Sciences, 2021, 545: 465–486. DOI: https://doi.org/10.1016/j.ins.2020.09.008.
ZHAO Hong, ZHAN Zhi-hui, ZHANG Jun. Adaptive guidance-based differential evolution with iterative feedback archive strategy for multimodal optimization problems [C]// IEEE Congress on Evolutionary Computation. Glasgow, United Kingdom: IEEE, 2020. DOI: https://doi.org/10.1109/CEC48606.2020.9185582.
CAO Yu-lian, ZHANG Han, LI Wen-feng, ZHOU Meng-chu, ZHANG Yu, CHAOVALITWONGSE W A. Comprehensive learning particle swarm optimization algorithm with local search for multimodal functions [J]. IEEE Transactions on Evolutionary Computation, 2019, 23(4): 718–731. DOI: https://doi.org/10.1109/TEVC.2018.2885075.
RIM C, PIAO S, LI Guo, PAK U. A niching chaos optimization algorithm for multimodal optimization [J]. Soft Computing, 2018, 22(2): 621–633. DOI: https://doi.org/10.1007/s00500-016-2360-2.
THIRUGNANASAMBANDAM K, PRAKASH S, SUBRAMANIAN V, POTHULA S, THIRUMAL V. Reinforced cuckoo search algorithm-based multimodal optimization [J]. Applied Intelligence, 2019, 49(6): 2059–2083. DOI: https://doi.org/10.1007/s10489-018-1355-3.
SUN Tao, ZHOU Zhi-hua. Structural diversity for decision tree ensemble learning [J]. Frontiers of Computer Science, 2018, 12(3): 560–570. DOI: https://doi.org/10.1007/s11704-018-7151-8.
DUBRULLE A A. Householder transformations revisited [J]. SIAM Journal on Matrix Analysis and Applications, 2000, 22(1): 33–40. DOI: https://doi.org/10.1137/S0895479898338561.
RODRIGUEZ-GALIANO V, SANCHEZ-CASTILLO M, CHICA-OLMO M, CHICA-RIVAS M. Machine learning predictive models for mineral prospectivity: an evaluation of neural networks, random forest, regression trees and support vector machines [J]. Ore Geology Reviews, 2015, 71(SI): 804–818. DOI: https://doi.org/10.1016/j.oregeorev.2015.01.001.
BREIMAN L, FRIEDMAN J H, OLSHEN R A, STONE C J. Classification and regression trees [M]. Belmont, CA: Wadsworth Advanced Books and Software, 1984. DOI: https://doi.org/10.1002/cyto.990080516.
CHEN Wei, XIE Xiao-shen, WANG Jia-le, PRADHAN B, HONG Hao-yuan, BUI D T, DUAN Zhao, MA Jian-quan. A comparative study of logistic model tree, random forest, and classification and regression tree models for spatial prediction of landslide susceptibility [J]. Catena, 2017, 151: 147–160. DOI: https://doi.org/10.1016/j.catena.2016.11.032.
DUA D, GRAFF C. UCI Machine learning repository [EB/OL]. [2020-01-13] http://archive.ics.uci.edu/ml,2017/.
COX N J, SCHECHTER C B. Speaking stata: how best to generate indicator or dummy variables [J]. Stata Journal, 2019, 19(1): 246–259. DOI: https://doi.org/10.1177/1536867X19830921.
JANEZ D. Statistical comparisons of classifiers over multiple data sets[J]. Journal of Machine Learning Research, 2006, 7: 1–30.
NISHIYAMA T, SEO T. The multivariate tukey-kramer multiple comparison procedure among four correlated mean vectors [J]. American Journal of Mathematical and Management Sciences, 2008, 28(1, 2): 115–130. DOI: https://doi.org/10.1080/01966324.2008.10737720.
KUNCHEVA L I. A bound on kappa-error diagrams for analysis of classifier ensembles [J]. IEEE Transactions on Knowledge and Data Engineering, 2013, 25(3): 494–501. DOI: https://doi.org/10.1109/TKDE.2011.234.
COHEN J. A coefficient of agreement for nominal scales [J]. Educational and Psychological Measurement, 1960, 20(1): 37–46. DOI: https://doi.org/10.1177/001316446002000104.
FLIGHT L, JULIOUS S A. The disagreeable behaviour of the kappa statistic [J]. Pharmaceutical Statistics, 2015, 14(1): 74–78. DOI: https://doi.org/10.1002/pst.1659.
Author information
Authors and Affiliations
Corresponding author
Additional information
Foundation item
Project(61603274) supported by the National Natural Science Foundation of China; Project(2017KJ249) supported by the Research Project of Tianjin Municipal Education Commission, China
Contributors
XU Zhe provided the concept, performed the experiments, and wrote the manuscript. NI Wei-chen helped with the data analysis and revision of the manuscript. JI Yue-hui contributed to manuscript preparation and the revision of the manuscript.
Conflict of interest
XU Zhe, NI Wei-chen and JI Yue-hui declare that they have no conflict of interest.
Rights and permissions
About this article
Cite this article
Xu, Z., Ni, Wc. & Ji, Yh. Rotation forest based on multimodal genetic algorithm. J. Cent. South Univ. 28, 1747–1764 (2021). https://doi.org/10.1007/s11771-021-4730-x
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11771-021-4730-x