|
|
Feature selection algorithm based on Levy flight |
ZHU Xiao-en, HAO Xin, XIA Shun-ren |
Key Laboratory of Biomedical Engineering of Ministry of Education, Zhejiang University, Hangzhou 310027, China |
|
|
Abstract A Levy flight random process based feature selection algorithm (LevyFS) was proposed in order to improve the speed of feature selection method. A multi-stages heuristic search strategy was defined during optimization process. Levy flight random process was introduced in local search behavior, and map between Levy flight distances and search operations was defined. During different search stages, map was used to change the probability of search behavior so as to control the direction and speed of local search behavior. Then local optimum was prevented. Experimental results show that LevyFS algorithm overcomes the limitation of heuristic methods and the average time cost of LevyFS algorithm is only one-third time cost of SFFS algorithm.
|
Published: 01 April 2013
|
|
基于Levy flight的特征选择算法
为了提高特征选择方法的计算速度,提出基于Levy flight随机过程的特征选择方法.该方法在寻优过程中定义基于启发式的分阶段搜索策略,在局部搜索行为中引入Levy flight随机过程,将Levy flight距离与搜索行为进行映射.在不同的搜索阶段,利用不同的映射区间改变搜索行为出现的概率,以该映射来控制局部搜索行为的方向和速度,从而避免了陷入局部最优的问题.实验结果表明,采用LevyFS算法克服了启发式特征选择方法的局限性,平均耗时仅为SFFS算法的1/3左右.
|
|
[1] GUYON I, ELISSEEFF A E. An introduction to variable and feature selection [J]. Journal of Machine Learning Research, 2003, 3(1): 1157-1182.
[2] AMALDI E, KANN V. On the approximability of minimizing nonzero variables or unsatisfied relations in linear systems [J]. Theoretical Computer Science, 1998, 209(1/2): 237-260.
[3] 王娟,慈林林,姚康泽.特征选择方法综述[J].计算机工程与科学,2005,27(12): 68-71.
WANG Juan, CI Lin-lin, YAO Kang-ze. Review of feature selection method [J]. Computer Engineering and Science, 2005, 27(12): 68-71.
[4] IL-SEOK O, JIN-SEON L, BYUNG-RO M. Hybrid genetic algorithms for feature selection [J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2004, 26(11): 1424-1437.
[5] WANG X, YANG J, TENG X, et al. Feature selection based on rough sets and particle swarm optimization [J]. Pattern Recognition Letters, 2007, 28(4): 459-471.
[6] REUNANEN J. Overfitting in making comparisons between variable selection methods [J]. Journal of Machine Learning Research, 2003, 3(1): 1371-1382.
[7] PAVLYUKEVICH I. Lévy flights, non-local search and simulated annealing [J]. Journal of Computational Physics, 2007, 226(2): 1830-1844.
[8] VISWANATHAN G M. Lévy flights and random searches [J]. Journal of Physics A: Mathematical and Theoretical, 2009, 42(43): 434003.
[9] BROWN C, LIEBOVITCH L, GLENDON R. Lévy flights in Dobe Ju/’hoansi foraging patterns [J]. Springer Netherlands, 2007, 35: 129-138.
[10] MANTEGNA R N, STANLEY H E. Stochastic process with ultraslow convergence to a Gaussian: the truncated Lévy flight [J]. Physical Review Letters, 1994, 73(22): 2946-2949.
[11] YANG X S. Nature-inspired metaheuristic algorithms [M]. 2nd ed. United Kingdom: Luniver Press, 2011: 1-6.
[12] ID P S A P. Introduction to feature selection toolbox 3: the C++ library for subset search, data modeling and classication [R]. Czech: University of Tennessee Institute of Agriculture, Czech Academy of Sciences, 2010. |
|
Viewed |
|
|
|
Full text
|
|
|
|
|
Abstract
|
|
|
|
|
Cited |
|
|
|
|
|
Shared |
|
|
|
|
|
Discussed |
|
|
|
|