Abstract
This chapter presents two real problems where feature selection has proved to be useful in improving performance. Section 5.1 is devoted to improving the classification accuracy of the KDD Cup 99 dataset, a benchmark in the intrusion detection field. A method based on the combination of discretization, filtering and classification algorithms is proposed, to be applied to both the binary and the multiclass version of the dataset. Then, the second problem is presented in Section 5.2, which is related to tear film lipid layer classification. A methodology making use of feature selection methods is proposed, achieving considerable improvements in performance. Since the second problem has the added handicap of having to reduce the cost associated with the features, in Section 5.3 we introduce a general framework for cost-based feature selection. Section 5.4 closes this chapter with a summary of its findings.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
Author information
Authors and Affiliations
Rights and permissions
Copyright information
© 2015 Springer International Publishing Switzerland
About this chapter
Cite this chapter
Bolón-Canedo, V., Sánchez-Maroño, N., Alonso-Betanzos, A. (2015). Application of Feature Selection to Real Problems. In: Feature Selection for High-Dimensional Data. Artificial Intelligence: Foundations, Theory, and Algorithms. Springer, Cham. https://doi.org/10.1007/978-3-319-21858-8_5
Download citation
DOI: https://doi.org/10.1007/978-3-319-21858-8_5
Publisher Name: Springer, Cham
Print ISBN: 978-3-319-21857-1
Online ISBN: 978-3-319-21858-8
eBook Packages: Computer ScienceComputer Science (R0)