Joint L2,1 Norm and Fisher Discrimination Constrained Feature Selection for Rational Synthesis of Microporous Aluminophosphates

Feature selection has been regarded as an effective tool to help researchers understand the generating process of data. For mining the synthesis mechanism of microporous AlPOs, this paper proposes a novel feature selection method by joint l2,1 norm and Fisher discrimination constraints (JNFDC). In order to obtain more effective feature subset, the proposed method can be achieved in two steps. The first step is to rank the features according to sparse and discriminative constraints. The second step is to establish predictive model with the ranked features, and select the most significant features in the light of the contribution of improving the predictive accuracy. To the best of our knowledge, JNFDC is the first work which employs the sparse representation theory to explore the synthesis mechanism of six kinds of pore rings. Numerical simulations demonstrate that our proposed method can select significant features affecting the specified structural property and improve the predictive accuracy. Moreover, comparison results show that JNFDC can obtain better predictive performances than some other state‐of‐the‐art feature selection methods.

[1]  Masashi Sugiyama,et al.  Feature Selection via L1-Penalized Squared-Loss Mutual Information , 2012, IEICE Trans. Inf. Syst..

[2]  Yvan Vander Heyden,et al.  Towards better understanding of feature-selection or reduction techniques for Quantitative Structure–Activity Relationship models , 2013 .

[3]  Shigeo Abe DrEng.  Pattern Classification , 2001, Springer London.

[4]  Jun Kong,et al.  A Novel Integrated Feature Selection Method for the Rational Synthesis of Microporous Aluminophosphate , 2012 .

[5]  X. Rosalind Wang,et al.  Human breath-print identification by E-nose, using information-theoretic feature selection prior to classification , 2015 .

[6]  David Zhang,et al.  Feature selection and analysis on correlated gas sensor data with recursive feature elimination , 2015 .

[7]  Jian Yang,et al.  Sparse discriminative feature selection , 2015, Pattern Recognit..

[8]  Jun Kong,et al.  Computational prediction of the formation of microporous aluminophosphates with desired structural features , 2010 .

[9]  Ruben H. Zamar,et al.  Exploiting Multiple Descriptor Sets in QSAR Studies , 2016, J. Chem. Inf. Model..

[10]  Apilak Worachartcheewan,et al.  Predictive QSAR modeling of aldose reductase inhibitors using Monte Carlo feature selection. , 2014, European journal of medicinal chemistry.

[11]  Simon C. K. Shiu,et al.  Unsupervised feature selection by regularized self-representation , 2015, Pattern Recognit..

[12]  Zhihui Lai,et al.  The L2, 1-norm-based unsupervised optimal feature selection with applications to action recognition , 2016, Pattern Recognit..

[13]  Li Shen,et al.  Cortical surface biomarkers for predicting cognitive outcomes using group l 2,1 norm , 2015, Neurobiology of Aging.

[14]  Masashi Sugiyama,et al.  High-Dimensional Feature Selection by Feature-Wise Kernelized Lasso , 2012, Neural Computation.

[16]  Huawen Liu,et al.  Fisher discrimination based low rank matrix recovery for face recognition , 2014, Pattern Recognit..

[17]  Harun Uzun,et al.  Prediction of gas storage capacities in metal organic frameworks using artificial neural network , 2015 .

[18]  Jihong Yu,et al.  Rational approaches toward the design and synthesis of zeolitic inorganic open-framework materials. , 2010, Accounts of chemical research.

[19]  Hong Yan,et al.  Robust classification using ℓ2, 1-norm based regression model , 2012, Pattern Recognit..

[20]  Nick C Fox,et al.  Gene-Wide Analysis Detects Two New Susceptibility Genes for Alzheimer's Disease , 2014, PLoS ONE.

[21]  Qiuqi Ruan,et al.  Sparse feature selection based on L2, 1/2-matrix norm for web image annotation , 2015, Neurocomputing.