An Innovative Feature Selection Using Fuzzy Entropy

In this paper, a new feature subset selection approach is introduced. The proposed approach consists of two phases. In the first phase, we tried to reduce the run time order of the algorithm which is critical for high dimensional datasets. In this phase, first entire dataset is classified and according to silhouette value, the best number of clusters in the dataset is found. Using this value, second, each feature is classified alone with the same cluster number and proposed entropy fuzzy measures for them are calculated. In the second phase, it is tried to find a feature subset that meets the boundaries to get a high accuracy degree. The proposed method is examined on different datasets. The examination results show that the proposed method leans to find and select the minimum number of features with negligible removing final classification accuracy, among different feature subset selection methods.

[1]  N. Chaikla,et al.  Genetic algorithms in feature selection , 1999, IEEE SMC'99 Conference Proceedings. 1999 IEEE International Conference on Systems, Man, and Cybernetics (Cat. No.99CH37028).

[2]  Gareth M. James,et al.  Functional linear discriminant analysis for irregularly sampled curves , 2001 .

[3]  I. Jolliffe.  Principal Component Analysis , 2002 .

[4]  S. Billings,et al.  Feature Subset Selection and Ranking for Data Dimensionality Reduction , 2007 .

[5]  Sankar K. Pal,et al.  Feature analysis: Neural network and fuzzy set theoretic approaches , 1997, Pattern Recognit..

[6]  Sankar K. Pal,et al.  Neuro-fuzzy feature evaluation with theoretical analysis , 1999, Neural Networks.

[7]  S T Roweis,et al.  Nonlinear dimensionality reduction by locally linear embedding. , 2000, Science.

[8]  Shyi-Ming Chen,et al.  Feature subset selection based on fuzzy entropy measures for handling classification problems , 2008, Applied Intelligence.

[9]  John C. Platt.  Using Analytic QP and Sparseness to Speed Training of Support Vector Machines , 1998, NIPS.

[10]  Roberto Battiti,et al.  Using mutual information for selecting features in supervised neural net learning , 1994, IEEE Trans. Neural Networks.

[11]  Xizhao Wang,et al.  OFFSS: optimal fuzzy-valued feature subset selection , 2003, IEEE Trans. Fuzzy Syst..

[12]  Johanna Smeyers-Verbeke,et al.  Chapter 33 - Supervised Pattern Recognition , 1998 .

[13]  Pierre Comon,et al.  Independent component analysis, A new concept? , 1994, Signal Process..

[14]  Eric O. Postma,et al.  Dimensionality Reduction: A Comparative Review , 2008 .

[15]  J. Tenenbaum,et al.  A global geometric framework for nonlinear dimensionality reduction. , 2000, Science.