File(s) under permanent embargo
Use of circle-segments as a data visualization technique for feature selection in pattern classification
chapter
posted on 2007-01-01, 00:00 authored by S Wang, C Loy, Chee Peng LimChee Peng Lim, W Lai, K TanOne of the issues associated with pattern classification using data based machine learning systems is the “curse of dimensionality”. In this paper, the circle-segments method is proposed as a feature selection method to identify important input features before the entire data set is provided for learning with machine learning systems. Specifically, four machine learning systems are deployed for classification, viz. Multilayer Perceptron (MLP), Support Vector Machine (SVM), Fuzzy ARTMAP (FAM), and k-Nearest Neighbour (kNN). The integration between the circle-segments method and the machine learning systems has been applied to two case studies comprising one benchmark and one real data sets. Overall, the results after feature selection using the circle segments method demonstrate improvements in performance even with more than 50% of the input features eliminated from the original data sets.
History
Title of book
Neural Information Processing 14th International Conference, ICONIP 2007, Kitakyushu, Japan, November 13-16, 2007, Revised Selected PapersSeries
Lecture notes in computer science ; 4984-4985.Chapter number
65Pagination
625 - 634Publisher
Springer-VerlagPlace of publication
Berlin, GermanyPublisher DOI
ISBN-13
9783540691549ISBN-10
3540691545Language
engPublication classification
B1.1 Book chapterCopyright notice
2007, SpringerExtent
116Editor/Contributor(s)
M IshikawaUsage metrics
Categories
No categories selectedKeywords
Licence
Exports
RefWorks
BibTeX
Ref. manager
Endnote
DataCite
NLM
DC