Selection of relevant variables for industrial process modeling by combining experimental data sensitivity and human knowledge

Selection of relevant variables from a high dimensional process operation setting space is a problem frequently encountered in industrial process modeling. This paper presents two global relevancy criteria, which permit to formalize and combine the sensitivity of experimental data and the conformity of human knowledge using a liner and a fuzzy model, respectively. The performances of these relevancy criteria and some well-known selection methods are compared through artificial and real datasets. The result validates the outperformance of fuzzy global relevancy criterion, especially when the number of learning data is small and noisy.

[1]  K. I. Ramachandran,et al.  Feature selection using Decision Tree and classification through Proximal Support Vector Machine for fault diagnostics of roller bearing , 2007 .

[2]  Richard Weber,et al.  A wrapper method for feature selection using Support Vector Machines , 2009, Inf. Sci..

[3]  Yinghua Lin,et al.  Nonlinear system input structure identification: two stage fuzzy curves and surfaces , 1998, IEEE Trans. Syst. Man Cybern. Part A.

[4]  Sanghamitra Bandyopadhyay,et al.  An efficient technique for superfamily classification of amino acid sequences: feature extraction, fuzzy clustering and prototype selection , 2005, Fuzzy Sets Syst..

[5]  Shigeo Abe,et al.  A novel approach to feature selection based on analysis of class regions , 1997, IEEE Trans. Syst. Man Cybern. Part B.

[6]  Xianyi Zeng,et al.  Designing Structural Parameters of Nonwovens Using Fuzzy Logic and Neural Networks , 2008 .

[7]  Yiu-ming Cheung,et al.  A new feature selection method for Gaussian mixture clustering , 2009, Pattern Recognit..

[8]  Yadong Wang,et al.  Improving fuzzy c-means clustering based on feature-weight learning , 2004, Pattern Recognit. Lett..

[9]  Witold Pedrycz,et al.  Effective classification using feature selection and fuzzy integration , 2008, Fuzzy Sets Syst..

[10]  Lei Liu,et al.  Feature selection with dynamic mutual information , 2009, Pattern Recognit..

[11]  Yong Qin,et al.  Research on input variable selection for numeric data based fuzzy modeling , 2003, Proceedings of the 2003 International Conference on Machine Learning and Cybernetics (IEEE Cat. No.03EX693).

[12]  Ebrahim H. Mamdani,et al.  An Experiment in Linguistic Synthesis with a Fuzzy Logic Controller , 1999, Int. J. Hum. Comput. Stud..

[13]  R. Fisher.  THE USE OF MULTIPLE MEASUREMENTS IN TAXONOMIC PROBLEMS , 1936 .

[14]  Gang Feng,et al.  Mamdani-type fuzzy controllers are universal fuzzy controllers , 2001, Fuzzy Sets Syst..

[15]  Yun Li,et al.  Feature selection based on loss-margin of nearest neighbor classification , 2009, Pattern Recognit..

[16]  Pat Langley,et al.  Selection of Relevant Features and Examples in Machine Learning , 1997, Artif. Intell..

[17]  Abraham Kandel,et al.  Information-theoretic algorithm for feature selection , 2001, Pattern Recognit. Lett..

[18]  U. Asan,et al.  A fuzzy approach to qualitative cross impact analysis , 2004 .

[19]  Pei-Chann Chang,et al.  A CBR-based fuzzy decision tree approach for database classification , 2010, Expert Syst. Appl..