Abstract Feature selection plays an important role in machine learning and data mining as a primary preprocessing step. A feature selection algorithm is presented based on fractal dimension and ant colony optimization algorithm. In this algorithm, fractal dimension is used as an evaluation mechanism and ant colony optimization algorithm is employed to accelerate the selection process. To evaluate the efficiency of the proposed algorithm, the SVM algorithm and K-fold cross validation are utilized to evaluate the classification accuracy on four datasets. The experimental results show the proposed algorithm can identify the better feature space with a great decrease of dataset dimension in a short time.
[1] Chen You, Cheng Xueqi, Li Yang, et al. Lightweight Intrusion Detection System Based on Feature Selection. Journal of Software, 2007, 18(7): 1639-1651 (in Chinese) (陈 友,程学旗,李 洋,等.基于特征选择的轻量级入侵检测系统.软件学报, 2007, 18(7): 1639-1651) [2] Jr Traina C, Wu L, Traina A, et al. Fast Feature Selection Using Fractal Dimension // Proc of the XV Brazilian Symposium on Databases. Paraiba, Brazil, 2000: 78-90 [3] Eneva E, Kumaraswamy K, Matteucci M. WEK WEK : A Study in Fractal Dimension and Dimensionality Reduction [EB/OL]. [2002- 7- 23]. http://home.dei.polimi.it/matteucc/Download/wekkem.pdf [4] Yan Guanghui. Performance Optimization of Fractal Dimensionality Reduction Algorithm. Journal of Lanzhou Jiaotong University, 2007, 26(1): 6-10,25 (in Chinese) (闫光辉.一种高效的分形属性选择算法.兰州交通大学学报, 2007, 26(1): 6-10,25) [5] Yan Guanghui, Li Zhanhuai, Yuan Liu.The Practical Method of Fractal Dimensionality Reduction Based on Z-Ordering Technique // Proc of the 2nd International Conference on Advanced Data Mining and Applications. Xi'an, China, 2006: 542-549 [6] Bao Yubin, Wang Zhuo, Sun Huanliang, et al. Fast Attribute Selection Algorithm Based on Fractal Dimension. Journal of Northeastern University: Natural Science, 2003, 24(6): 527-530 (in Chinese) (鲍玉斌,王 琢,孙焕良,等.一种基于分形维的快速属性选择算法.东北大学学报:自然科学版, 2003, 24(6): 527-530) [7] Zhang Chunkai, Hu Hong. Feature Selection Using the Hybrid of Ant Colony Optimization and Mutual Information for the Forecaster // Proc of the International Conference on Machine Learning and Cybernetics. Guangzhou, China, 2005, Ⅲ: 1728-1732 [8] Ahmed A A. Ant Colony Optimization for Feature Subset Selection. Proc of World Academy of Science, Engineering and Technology, 2005, 4: 35-38 [9] Caballero Y, Delia A, Bello R, et al. Feature Selection Algorithms Using Rough Set Theory // Proc of the 7th International Conference on Intelligent Systems Design and Applications. Piscataway, USA, 2007: 407-411 [10] Lee H D, Monard M C, Wu F C. A Fractal Dimension Based Filter Algorithm to Select Features for Supervised Learning // Proc of the Ibero-American Conference on Artificial Intelligence. Ribeiro Preto, Brazil, 2006: 278-288 [11] Ye Zhiwei, Zheng Zhaobao, Yu Xin. A Novel Approach to Feature Selection Based on Ant Colony Optimization Algorithm Hybridization of Heuristic Information. Science and Technology Information of Surveying and Mapping, 2007, 3: 5-12 (in Chinese) (叶志伟,郑肇葆,虞 欣.融和启发信息的蚁群特征选择方法.测绘科技情报, 2007, 3: 5-12) [12] Witten I H, Frank E. Practical Machine Learning Tools and Techniques. 2nd Edition. Orlando, USA: Morgan Kaufmann Publishers, 2005 [13] Hall M A. Correlation-Based Feature Selection for Discrete and Numeric Class Machine Learning // Proc of the 17th International Conference on Machine Learning. Stanford, USA, 2000: 359-366 [14] Lee H D, Monard M C, Wu F C. Feature Subset Selection for Supervised Learning Using Fractal Dimension // Proc of the 5th Congress of Logic Applied to Technology. Hyogo, Japan, 2005: 1-8 [15] Chang C C, Lin C J. LIBSVM—A Library for Support Vector Machines [DB/OL]. [2008- 01- 10]. http://www.csie.ntu.edu.tw/~cjlin/libsvm/ [16] Zhang Haiqin, Perng Changshing, Cai Qingsheng. An Improved Algorithm for Feature Selection Using Fractal Dimension [EB/OL]. [2008- 01- 10]. http://ftp.informatik.rwth-aachen.de/Publications/CEUR-WS/Vol-124/06zhang.pdf
2009, Vol. 22 
