Definition
Evolutionary feature selection and construction (EFSC) is a bio-inspired methodology for explicit modification of input data of a learning system. It uses evolutionary computation (EC) to find a mapping from the original data representation space onto a secondary representation space. In evolutionary feature selection (EFS), that mapping consists in dropping off some of the features ( attributes) from the original representation, so the dimensionality of the resulting representation space is not greater than that of the original space. In evolutionary feature construction (EFC), evolutionary algorithm creates (synthesizes) new features (derived attributes) that complement and/or replace the original ones. Therefore, EFS may be considered as special case of EFC.
A typical EFSC algorithm maintains a...
Recommended Reading
Bhanu, B., Lin, Y., & Krawiec, K. (2005). Evolutionary synthesis of pattern recognition systems. New York: Springer-Verlag.
Howard, D., Roberts, S. C., & Ryan, C. (2006). Pragmatic genetic programming strategy for the problem of vehicle detection in airborne reconnaissance. Pattern Recognition Letters, 27(11), 1275–1288.
Jaśkowski, W., Krawiec, K., & Wieloch, B. (2007). Knowledge reuse in genetic programming applied to visual learning. In Dirk Thierens et al. (eds.), GECCO ’07: In Proceedings of the 9th annual conference on Genetic and evolutionary computation, (Vol 2, pp. 1790–1797), London, 2007. ACM Press.
Komosiński, M., & Krawiec, K. (2000). Evolutionary weighting of image features for diagnosing of CNS tumors. Artificial Intelligence in Medicine, 19(1), 25–38.
Krawiec, K., & Bhanu, B. (2005). Visual learning by coevolutionary feature synthesis. IEEE Transactions on System, Man, and Cybernetics – Part B, 35(3), 409–425.
Krawiec, K., Howard, D., & Zhang, M. (2007). Overview of object detection and image analysis by means of genetic programming techniques. In Proceedings of frontiers in the convergence of bioscience and information technologies 2007 (fbit2007), Jeju, Korea, october 11–13, 2007 (pp. 779–784). IEEE CS Press.
Langdon, W., Gustafson, S., & Koza, J. (2009). The genetic programming bibliography. ([online] http://www.cs.bham.ac.uk/~wbl/biblio/)
Neshatian, K., & Zhang, M. (2009). Genetic programming for feature subset ranking in binary classification problems. In L. Vanneschi, S. Gustafson, A. Moraglio, I. vanoe De Falco, & M. Ebner (Eds.), Genetic programming (pp. 121–132). Springer.
Puente, C., Olague, G., Smith, S. V., Bullock, S. H., González-Botello, M. A., & Hinojosa-Corona, A. (2009). A novel GP approach to synthesize vegetation indices for soil eros ion assessment. In M. Giacobini et al. (Eds.), Applications of evolutionary computing (pp. 375–384). Springer.
Quintana, M. I., Poli, R., & Claridge, E. (2006) Morphological algorithm design for binary images using genetic programming. Genetic Programming and Evolvable Machines, 7(1), 81–102.
Rizki, M. M., Zmuda, M. A., & Tamburino, L. A. (2002). Evolving pattern recognition systems. IEEE Transactions on Evolutionary Computation, 6(6), 594–609.
Teller, A., & Veloso, M. (1997). PADO: A new learning architecture for object recognition. In K. Ikeuchi & M. Veloso (Eds.), Symbolic visual learning (pp. 77–112). New York: Oxford Press.
Vafaie, H., & Imam, I. F. (1994). Feature selection methods: genetic algorithms vs. greedy-like search. In Proceedings of international conference on fuzzy and intelligent control systems.
Yang, J., & Honavar, V. (1998). Feature subset selection using a genetic algorithm. IEEE Transactions on Intelligent Systems, 13(2), 44–49.
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2011 Springer Science+Business Media, LLC
About this entry
Cite this entry
Krawiec, K. (2011). Evolutionary Feature Selection and Construction. In: Sammut, C., Webb, G.I. (eds) Encyclopedia of Machine Learning. Springer, Boston, MA. https://doi.org/10.1007/978-0-387-30164-8_279
Download citation
DOI: https://doi.org/10.1007/978-0-387-30164-8_279
Publisher Name: Springer, Boston, MA
Print ISBN: 978-0-387-30768-8
Online ISBN: 978-0-387-30164-8
eBook Packages: Computer ScienceReference Module Computer Science and Engineering