Publication: Dual problems: Attribute selection and example selection
Issued Date
2004-12-01
Resource Type
Other identifier(s)
2-s2.0-12744274418
Rights
Mahidol University
Rights Holder(s)
SCOPUS
Bibliographic Citation
Proceedings of the International Conference on Internet Computing, IC'04. Vol.1, (2004), 60-65
Suggested Citation
Puntip Pattaraintakom, Kanlaya Naruedomkul, Nick Gereone Dual problems: Attribute selection and example selection. Proceedings of the International Conference on Internet Computing, IC'04. Vol.1, (2004), 60-65. Retrieved from: https://repository.li.mahidol.ac.th/handle/123456789/21315
Research Projects
Organizational Units
Authors
Journal Issue
Thesis
Title
Dual problems: Attribute selection and example selection
Other Contributor(s)
Abstract
We proposed an alternative methodology, Dual AE, for attribute selection and example selection. The Dual AE is designed to increase the accuracy and efficiency of attribute selection and example selection. The central idea behind the Dual AE is to generate the attribute candidates by "Attribute candidate generation" after the data were cleaned in "Data pre-processing module". All the candidates are weighted in the "Attribute weighting" so that the most informative attribute set can be selected in "Clustering". The accuracy of the selected attribute set is verified in the last module, "Validation". To demonstrate the Dual AE method, we have implemented it on the prototypical data set. In this paper we present the intuitions behind the design of Dual AE and some experimental results.