Attribute Selection for Numerical Databases that Contain Correlations
Received:October 14, 2008  Revised:December 19, 2008  Download PDF
Taufik Djatna,Yasuhiko Morimoto. Attribute Selection for Numerical Databases that Contain Correlations. International Journal of Software and Informatics, 2008,2(2):125~139
Hits: 5632
Download times: 7009
Taufik Djatna  Yasuhiko Morimoto
Fund:This work was supported by KAKENHI (#19500123). Tau k Djatna was supported by scholarship of MEXT Japan.
Abstract:There are many correlated attributes in a database. Conventional attribute selection methods are not able to handle such correlations and tend to eliminate important rules that exist in correlated attributes. In this paper, we propose an attribute selection method that preserves important rules on correlated attributes. We rst compute a ranking of attributes by using conventional attribute selection methods. In addition, we compute two-dimensional rules for each pair of attributes and evaluate their importance for predicting a target attribute. Then, we evaluate the shapes of important two-dimensional rules to pick up hidden important attributes that are under-estimated by conventional attribute selection methods. After the shape evaluation, we re-calculate the ranking so that we can preserve the important correlations. Intensive experiments show that the proposed method can select important correlated attributes that are eliminated by conventional methods.
keywords:Feature Selection  Correlated Attribute  Two Dimensional Rule  Region Shape
View Full Text  View/Add Comment  Download reader



Top Paper  |  FAQ  |  Guest Editors  |  Email Alert  |  Links  |  Copyright  |  Contact Us

© Copyright by Institute of Software, the Chinese Academy of Sciences

京公网安备 11040202500065号