{"title":"Genetic Programming Based Data Projections for Classification Tasks","authors":"C\u00e9sar Est\u00e9banez, Ricardo Aler, Jos\u00e9 M. Valls","volume":7,"journal":"International Journal of Computer and Information Engineering","pagesStart":2195,"pagesEnd":2201,"ISSN":"1307-6892","URL":"https:\/\/publications.waset.org\/pdf\/15252","abstract":"<p>In this paper we present a GP-based method for automatically evolve projections, so that data can be more easily classified in the projected spaces. At the same time, our approach can reduce dimensionality by constructing more relevant attributes. Fitness of each projection measures how easy is to classify the dataset after applying the projection. This is quickly computed by a Simple Linear Perceptron. We have tested our approach in three domains. The experiments show that it obtains good results, compared to other Machine Learning approaches, while reducing dimensionality in many cases.<\/p>\r\n","references":"[1] N. Cristianini and J. Shawe-Taylor. An introduction to Support Vector\r\nMachines (and other kernel-based learning methods). Cambridge\r\nUniversity Press, 2000.\r\n[2] T. Fawcett and P. Utgoff. A hybrid method for feature generation. In\r\nProceedings of the Eighth International Workshop on Machine Learning,\r\npages 137- 141, Evanston, IL.\r\n[3] S. Kramer. Cn2-mci: A two-step method for constructive induction. In\r\nProceedings of ML-COLT-94.\r\n[4] B. Pfahringer. Cipf 2.0: A robust constructive induction system. In\r\nProceedings of ML-COLT-94, 1994. W. D. Doyle, \"Magnetization\r\nreversal in films with biaxial anisotropy,\" in 1987 Proc. INTERMAG\r\nConf., pp. 2.2-1-2.2-6. G. W. Juette and L. E. Zeffanella, \"Radio noise\r\ncurrents n short sections on bundle conductors (Presented Conference\r\nPaper style),\" presented at the IEEE Summer power Meeting, Dallas,\r\nTX, June 22-27, 1990, Paper 90 SM 690-0 PWRS.\r\n[5] John R. Koza. Genetic Programming: On the Programming of\r\nComputers by Means of Natural Selection. MIT Press, Cambridge, MA,\r\nUSA, 1992.\r\n[6] John R. Koza. Genetic Programming II: Automatic Discovery of\r\nReusable Programs. MIT Press, Cambridge Massachusetts, May 1994.\r\n[7] B.D. Ripley. Pattern Recognition and Neural Networks. Cambridge:\r\nCambridge University Press, 1996.\r\n[8] D. Michie, D. J. Spiegelhalter, and C.C. Taylor. Machine learning,\r\nneural and statistical classification. Ellis Horwood, 1994.\r\n[9] Benjamin Blankertz, Gabriel Curio, and Klaus-Robert M\u252c\u00bfuller.\r\nClassifying single trial eeg: Towards brain computer interfacing. In\r\nAdvances in Neural Inf. Proc. Systems 14 (NIPS 01), 2002.\r\n[10] Fernando E. B. Otero, Monique M. S. Silva, Alex A. Freitas, and Julio\r\nC. Nievola. Genetic programming for attribute construction in data\r\nmining. In Conor Ryan, Terence Soule, Maarten Keijzer, Edward Tsang,\r\nRiccardo Poli, and Ernesto Costa, editors, Genetic Programming,\r\nProceedings of EuroGP-2003, volume 2610 of LNCS, pages 389-398,\r\nEssex, 14-16 April 2003. Springer-Verlag.\r\n[11] Krzysztof Krawiec. Genetic programming-based construction of features\r\nfor machine learning and knowledge discovery tasks. Genetic\r\nProgramming and Evolvable Machines, 3(4):329-343, December 2002.\r\n[12] Tom Howley and Michael G. Madden. The genetic kernel support\r\nvector machine: Description and evaluation. Artificial Intelligence\r\nReview, To appear, 2005.\r\n[13] S. Davis S. Perkins J. Ma R. Porter D. Eads, D. Hill and J. Theiler.\r\nGenetic algorithms and support vector machines for time series\r\nclassification. In Proceedings SPIE 4787 Conference on Visualization\r\nand Data Analysis, pages 74-85, 2002.\r\n[14] John J. Szymanski, Steven P. Brumby, Paul Pope, Damian Eads, Diana\r\nEsch-Mosher, Mark Galassi, Neal R. Harvey, Hersew D. W. McCulloch,\r\nSimon J. Perkins, Reid Porter, James Theiler, A. Cody Young, Jeffrey J.\r\nBloch, and Nancy David. Feature extraction from multiple data sources\r\nusing genetic programming. In Sylvia S. Shen and Paul E. Lewis,\r\neditors, Algorithms and Technologies for Multispectral, Hyperspectral,\r\nand Ultraspectral Imagery VIII, volume 4725 of SPIE, pages 338-345,\r\nAugust 2002.\r\n[15] Neal R. Harvey, James Theiler, Steven P. Brumby, Simon Perkins, John\r\nJ. Szymanski, Jeffrey J. Bloch, Reid B. Porter, Mark Galassi, and A.\r\nCody Young. Comparison of GENIE and conventional supervised\r\nclassifiers for multispectral image feature extraction. IEEE Transactions\r\non Geoscience and Remote Sensing, 40(2):393-404, February 2002.","publisher":"World Academy of Science, Engineering and Technology","index":"Open Science Index 7, 2007"}