Logistic Regression Based on Statistical Learning Model with Linearized Kernel for Classification

In this paper, we propose a logistic regression classification method based on the integration of statistical learning model with linearized kernel pre-processing. The single Gaussian and fusion cosine kernels are adopted for pre-processing respectively. models generalized linear additive model. Using model, elastic net regularization is to explore grouping effect feature space. an overlap group-lasso penalty used fit sparse functions within Experiment results Extended Yale-B face database AR demonstrate effectiveness proposed method. improved solution also efficiently obtained using our spectra data.