Feature Selection based on Importance and Interaction Indexes - Hierarchical Fuzzy Rule Classifier Application

This paper proposed an extension of an iterative method to select suitable features for pattern recognition context. The main improvement is to replace its iterative step with another criterion based on importance and interaction indexes, providing suitable feature reduced set. This new scheme is embedded on a hierarchical fuzzy rule classification system. At last, each node gathers a set of classes having a similar aspect. The aim of the proposed method is to automatically extract an efficient subset of suitable features for each node. A selection of features is given. The associated criterion is directly based on importance index and assessment of positive and negative interaction between features. An experimental study, made in a wood defect recognition industrial context, shows the proposed method is efficient to producing significantly fewer rules.

[1]  Laurent Wendling,et al.  Extraction of Consistent Subsets of Descriptors using Choquet Integral , 2006, 18th International Conference on Pattern Recognition (ICPR'06).

[2]  Manuele Bicego,et al.  Statistical classification of raw textile defects , 2004, ICPR 2004.

[3]  Vincent Bombardier,et al.  Fuzzy rule classifier: Capability for generalization in wood color recognition , 2010, Eng. Appl. Artif. Intell..

[4]  M. Grabisch.  Fuzzy integral in multicriteria decision making , 1995 .

[5]  L. S. Shapley,et al.  17. A Value for n-Person Games , 1953 .

[6]  H.M. Abbas,et al.  Performance of neural classifiers for fabric faults classification , 2005, Proceedings. 2005 IEEE International Joint Conference on Neural Networks, 2005..

[7]  Hongbin Zhang,et al.  Feature selection using tabu search method , 2002, Pattern Recognit..

[8]  Laurent Wendling,et al.  Improving Fuzzy Rule Classifier by Extracting Suitable Features From Capacities With Respect to the Choquet Integral , 2008, IEEE Transactions on Systems, Man, and Cybernetics, Part B (Cybernetics).

[9]  Michel Grabisch,et al.  Classification by fuzzy integral: performance and tests , 1994, CVPR 1994.

[10]  Xuezhi Yang,et al.  Fabric defect classification using wavelet frames and minimum classification error training , 2002, Conference Record of the 2002 IEEE Industry Applications Conference. 37th IAS Annual Meeting (Cat. No.02CH37344).

[11]  Josef Kittler,et al.  Floating search methods in feature selection , 1994, Pattern Recognit. Lett..

[12]  Mineichi Kudo,et al.  Comparison of algorithms that select features for pattern classifiers , 2000, Pattern Recognit..