Feature selection based on intuitionistic hesitant fuzzy regularized LASSO regression
Yijin Zhang, Jie Huang, Miao Luo, Shengxia Tu · 2022
Excellent feature selection methods can reduce the data dimensionality and improve the efficiency of machine learning tasks. Logistic regression model is one of the models that are widely used for feature selection. In this paper, we propose a regularized LASSO logistic regression model by intuitionistic hesitant fuzzy correlation coefficients, which introduces fuzzy information into the logistic regression model to further enhance its feature selection capability. We design experiments to verify the effectiveness of the feature selection method proposed in this paper. The experimental results show that the method can perform feature selection effectively, and the selected variables can complete the classification task accurately.