{"title":"A Fuzzy-Rough Feature Selection Based on Binary Shuffled Frog Leaping Algorithm","authors":"Javad Rahimipour Anaraki, Saeed Samet, Mahdi Eftekhari, Chang Wook Ahn","volume":141,"journal":"International Journal of Computer and Information Engineering","pagesStart":722,"pagesEnd":730,"ISSN":"1307-6892","URL":"https:\/\/publications.waset.org\/pdf\/10009513","abstract":"Feature selection and attribute reduction are crucial
\r\nproblems, and widely used techniques in the field of machine
\r\nlearning, data mining and pattern recognition to overcome the
\r\nwell-known phenomenon of the Curse of Dimensionality. This paper
\r\npresents a feature selection method that efficiently carries out attribute
\r\nreduction, thereby selecting the most informative features of a dataset.
\r\nIt consists of two components: 1) a measure for feature subset
\r\nevaluation, and 2) a search strategy. For the evaluation measure,
\r\nwe have employed the fuzzy-rough dependency degree (FRFDD)
\r\nof the lower approximation-based fuzzy-rough feature selection
\r\n(L-FRFS) due to its effectiveness in feature selection. As for the
\r\nsearch strategy, a modified version of a binary shuffled frog leaping
\r\nalgorithm is proposed (B-SFLA). The proposed feature selection
\r\nmethod is obtained by hybridizing the B-SFLA with the FRDD. Nine
\r\nclassifiers have been employed to compare the proposed approach
\r\nwith several existing methods over twenty two datasets, including
\r\nnine high dimensional and large ones, from the UCI repository.
\r\nThe experimental results demonstrate that the B-SFLA approach
\r\nsignificantly outperforms other metaheuristic methods in terms of the
\r\nnumber of selected features and the classification accuracy.","references":"[1] X. Zhao, D. Li, B. Yang, C. Ma, Y. Zhu, and H. Chen, \u201cFeature selection\r\nbased on improved ant colony optimization for online detection of\r\nforeign fiber in cotton,\u201d Applied Soft Computing, vol. 24, pp. 585 \u2013\r\n596, 2014. [2] E. Hancer, B. Xue, D. Karaboga, and M. Zhang, \u201cA binary {ABC}\r\nalgorithm based on advanced similarity scheme for feature selection,\u201d\r\nApplied Soft Computing, vol. 36, pp. 334 \u2013 348, 2015.\r\n[3] N. Sreeja and A. Sankar, \u201cPattern matching based classification using ant\r\ncolony optimization based feature selection,\u201d Applied Soft Computing,\r\nvol. 31, pp. 91 \u2013 102, 2015.\r\n[4] S. Saha, R. Spandana, A. Ekbal, and S. Bandyopadhyay, \u201cSimultaneous\r\nfeature selection and symmetry based clustering using multiobjective\r\nframework,\u201d Applied Soft Computing, vol. 29, pp. 479 \u2013 486, 2015.\r\n[5] X. Han, \u201cImplicit feature selection for omics data phenotype\r\ndiscrimination,\u201d Applied Soft Computing, vol. 20, pp. 70 \u2013 82, 2014,\r\nhybrid intelligent methods for health technologies.\r\n[6] S.-W. Lin, K.-C. Ying, C.-Y. Lee, and Z.-J. Lee, \u201cAn intelligent\r\nalgorithm with feature selection and decision rules applied to anomaly\r\nintrusion detection,\u201d Applied Soft Computing, vol. 12, no. 10, pp. 3285\r\n\u2013 3290, 2012.\r\n[7] A. M. Canuto, K. M. Vale, A. Feitos, and A. Signoretti, \u201cReinsel: A\r\nclass-based mechanism for feature selection in ensemble of classifiers,\u201d\r\nApplied Soft Computing, vol. 12, no. 8, pp. 2517 \u2013 2529, 2012.\r\n[8] K. Manimala, K. Selvi, and R. Ahila, \u201cHybrid soft computing techniques\r\nfor feature selection and parameter optimization in power quality data\r\nmining,\u201d Applied Soft Computing, vol. 11, no. 8, pp. 5485 \u2013 5497, 2011.\r\n[9] R. Nock and M. Sebban, \u201cSharper bounds for the hardness of prototype\r\nand feature selection,\u201d in Algorithmic Learning Theory, ser. Lecture\r\nNotes in Computer Science, H. Arimura, S. Jain, and A. Sharma, Eds.\r\nSpringer Berlin Heidelberg, 2000, vol. 1968, pp. 224\u2013238.\r\n[10] S. C. Yusta, \u201cDifferent metaheuristic strategies to solve the feature\r\nselection problem,\u201d Pattern Recognition Letters, vol. 30, no. 5, pp. 525\r\n\u2013 534, 2009.\r\n[11] P. Pudil, J. Novoviov, and P. Somol, \u201cFeature selection toolbox software\r\npackage,\u201d Pattern Recognition Letters, vol. 23, no. 4, pp. 487 \u2013 492,\r\n2002.\r\n[12] M. ElAlami, \u201cA filter model for feature subset selection based on genetic\r\nalgorithm,\u201d Knowledge-Based Systems, vol. 22, no. 5, pp. 356 \u2013 362,\r\n2009.\r\n[13] S. Nemati, M. E. Basiri, N. Ghasem-Aghaee, and M. H. Aghdam, \u201cA\r\nnovel acoga hybrid algorithm for feature selection in protein function\r\nprediction,\u201d Expert Systems with Applications, vol. 36, no. 10, pp. 12 086\r\n\u2013 12 094, 2009.\r\n[14] S. M. Vieira, J. M. Sousa, and T. A. Runkler, \u201cTwo cooperative ant\r\ncolonies for feature selection using fuzzy models,\u201d Expert Systems with\r\nApplications, vol. 37, no. 4, pp. 2714 \u2013 2723, 2010.\r\n[15] M. Sebban and R. Nock, \u201cA hybrid filter\/wrapper approach of feature\r\nselection using information theory,\u201d Pattern Recognition, vol. 35, no. 4,\r\npp. 835 \u2013 846, 2002.\r\n[16] K. Thangavel and A. Pethalakshmi, \u201cDimensionality reduction based on\r\nrough set theory: A review,\u201d Applied Soft Computing, vol. 9, no. 1, pp.\r\n1 \u2013 12, 2009.\r\n[17] A. Verikas, M. Bacauskiene, D. Valincius, and A. Gelzinis, \u201cPredictor\r\noutput sensitivity and feature similarity-based feature selection,\u201d Fuzzy\r\nSets and Systems, vol. 159, no. 4, pp. 422 \u2013 434, 2008.\r\n[18] C. Degang and Z. Suyun, \u201cLocal reduction of decision system with fuzzy\r\nrough sets,\u201d Fuzzy Sets and Systems, vol. 161, no. 13, pp. 1871 \u2013 1883,\r\n2010.\r\n[19] R. Jensen and Q. Shen, \u201cNew approaches to fuzzy-rough feature\r\nselection,\u201d Fuzzy Systems, IEEE Transactions on, vol. 17, no. 4, pp.\r\n824\u2013838, Aug 2009.\r\n[20] Y. Chen, D. Miao, and R. Wang, \u201cA rough set approach to feature\r\nselection based on ant colony optimization,\u201d Pattern Recognition Letters,\r\nvol. 31, no. 3, pp. 226 \u2013 233, 2010.\r\n[21] N. Suguna and K. Thanushkodi, \u201cA novel rough set reduct algorithm\r\nfor medical domain based on bee colony optimization,\u201d Journal of\r\nComputing, vol. 2, no. 6, pp. 49\u201354, June 2010.\r\n[22] X. Wang, J. Yang, X. Teng, W. Xia, and R. Jensen, \u201cFeature\r\nselection based on rough sets and particle swarm optimization,\u201d Pattern\r\nRecognition Letters, vol. 28, no. 4, pp. 459 \u2013 471, 2007.\r\n[23] J. Wr\u00b4oblewski, \u201cFinding minimal reducts using genetic algorithms,\u201d\r\nin Proccedings of the second annual join conference on infromation\r\nscience, 1995, pp. 186\u2013189.\r\n[24] J. R. Anaraki and M. Eftekhari, \u201cRough set based feature selection:\r\nA review,\u201d in Information and Knowledge Technology (IKT), 2013 5th\r\nConference on, May 2013, pp. 301\u2013306.\r\n[25] R. Jensen and Q. Shen, \u201cFuzzy-rough data reduction with ant colony\r\noptimization,\u201d Fuzzy Sets and Systems, vol. 149, no. 1, pp. 5 \u2013 20,\r\n2005.\r\n[26] J. Xiang, X. Han, F. Duan, Y. Qiang, X. Xiong, Y. Lan, and\r\nH. Chai, \u201cA novel hybrid system for feature selection based on an\r\nimproved gravitational search algorithm and k-nn method,\u201d Applied Soft\r\nComputing, vol. 31, pp. 293 \u2013 307, 2015.\r\n[27] S. M. Vieira, L. F. Mendona, G. J. Farinha, and J. M. Sousa, \u201cModified\r\nbinary {PSO} for feature selection using {SVM} applied to mortality\r\nprediction of septic patients,\u201d Applied Soft Computing, vol. 13, no. 8,\r\npp. 3494 \u2013 3504, 2013.\r\n[28] Z. Xu, G. Huang, K. Q. Weinberger, and A. X. Zheng, \u201cGradient\r\nboosted feature selection,\u201d in Proceedings of the 20th ACM SIGKDD\r\ninternational conference on Knowledge discovery and data mining.\r\nACM, 2014, pp. 522\u2013531.\r\n[29] J. H. Friedman, \u201cGreedy function approximation: a gradient boosting\r\nmachine,\u201d Annals of statistics, pp. 1189\u20131232, 2001.\r\n[30] L. Breiman, J. Friedman, C. J. Stone, and R. A. Olshen, Classification\r\nand regression trees. CRC press, 1984.\r\n[31] Z. Pawlak, \u201cRough sets,\u201d International Journal of Computer &\r\nInformation Sciences, vol. 11, no. 5, pp. 341\u2013356, 1982.\r\n[32] J. Komorowski, Z. Pawlak, L. Polkowski, and A. Skowron, \u201cRough sets:\r\nA tutorial,\u201d in Rough-Fuzzy Hybridization: A New Trend in Decision\r\nMaking, S. K. Pal and A. Skowron, Eds. Secaucus, NJ, USA:\r\nSpringer-Verlag New York, Inc., 1998, pp. 3\u201398.\r\n[33] M. Eusuff, K. Lansey, and F. Pasha, \u201cShuffled frog-leaping algorithm:\r\na memetic meta-heuristic for discrete optimization,\u201d Engineering\r\nOptimization, vol. 38, no. 2, pp. 129\u2013154, 2006.\r\n[34] Q. Duan, S. Sorooshian, and V. Gupta, \u201cEffective and efficient global\r\noptimization for conceptual rainfall-runoff models,\u201d Water Resources\r\nResearch, vol. 28, no. 4, pp. 1015\u20131031, 1992.\r\n[35] J. Kennedy and R. Eberhart, \u201cParticle swarm optimization,\u201d in Neural\r\nNetworks, 1995. Proceedings., IEEE International Conference on, vol. 4,\r\nNov 1995, pp. 1942\u20131948 vol.4.\r\n[36] A. S. Reddy and K. Vaisakh, \u201cEnvironmental constrained economic\r\ndispatch by modified shuffled frog leaping algorithm,\u201d Journal of\r\nBioinformatics and Intelligent Control, vol. 2, no. 3, pp. 216\u2013222, 2013.\r\n[37] A. M. Radzikowska and E. E. Kerre, \u201cA comparative study of fuzzy\r\nrough sets,\u201d Fuzzy Sets and Systems, vol. 126, no. 2, pp. 137 \u2013 155,\r\n2002.\r\n[38] S. Kamyab, M. Eftekhari, and J. R. Anaraki, \u201cA novel rough set based\r\ndissimilarity measure and its application in multimodal optimization,\u201d\r\nin Artificial Intelligence and Signal Processing (AISP), 2012 16th CSI\r\nInternational Symposium on, May 2012, pp. 180\u2013185.\r\n[39] R. Jensen and Q. Shen, Computational intelligence and feature selection:\r\nrough and fuzzy approaches. John Wiley & Sons, 2008, vol. 8.\r\n[40] M. Lichman, \u201cUCI machine learning repository,\u201d 2013. (Online).\r\nAvailable: http:\/\/archive.ics.uci.edu\/ml.\r\n[41] A. Tsanas, M. A. Little, C. Fox, and L. O. Ramig, \u201cObjective automatic\r\nassessment of rehabilitative speech treatment in parkinson\u2019s disease,\u201d\r\nNeural Systems and Rehabilitation Engineering, IEEE Transactions on,\r\nvol. 22, no. 1, pp. 181\u2013190, 2014.\r\n[42] B. A. Johnson, \u201cHigh-resolution urban land-cover classification using a\r\ncompetitive multi-scale object-based approach,\u201d Remote Sensing Letters,\r\nvol. 4, no. 2, pp. 131\u2013140, 2013.\r\n[43] B. Johnson and Z. Xie, \u201cClassifying a high resolution image of an urban\r\narea using super-object information,\u201d ISPRS Journal of Photogrammetry\r\nand Remote Sensing, vol. 83, pp. 40\u201349, 2013.\r\n[44] P. Van Der Putten and M. van Someren, \u201cCoil challenge 2000: The\r\ninsurance company case,\u201d Published by Sentient Machine Research,\r\nAmsterdam. Also a Leiden Institute of Advanced Computer Science\r\nTechnical Report, vol. 9, pp. 1\u201343, 2000.\r\n[45] I. Guyon, S. Gunn, A. Ben-Hur, and G. Dror, \u201cResult analysis of the nips\r\n2003 feature selection challenge,\u201d in Advances in neural information\r\nprocessing systems, 2004, pp. 545\u2013552.\r\n[46] M. Hall, E. Frank, G. Holmes, B. Pfahringer, P. Reutemann, and I. H.\r\nWitten, \u201cThe weka data mining software: An update,\u201d SIGKDD Explor.\r\nNewsl., vol. 11, no. 1, pp. 10\u201318, Nov. 2009.\r\n[47] K.-C. Wong, C.-H. Wu, R. K. Mok, C. Peng, and Z. Zhang,\r\n\u201cEvolutionary multimodal optimization using the principle of locality,\u201d\r\nInformation Sciences, vol. 194, pp. 138 \u2013 170, 2012.","publisher":"World Academy of Science, Engineering and Technology","index":"Open Science Index 141, 2018"}