Abstract:Fuzzy rough entropy established in the notion of fuzzy rough set theory, which has been effectively and efficiently applied for feature selection to handle the uncertainty in real-valued datasets. Further, Fuzzy rough mutual information has been presented by integrating information entropy with fuzzy rough set to measure the importance of features. However, none of the methods till date can handle noise, uncertainty and vagueness simultaneously due to both judgement and identification, which lead to degrade the overall performances of the learning algorithms with the increment in the number of mixed valued conditional features. In the current study, these issues are tackled by presenting a novel intuitionistic fuzzy (IF) assisted mutual information concept along with IF granular structure. Initially, a hybrid IF similarity relation is introduced. Based on this relation, an IF granular structure is introduced. Then, IF rough conditional and joint entropies are established. Further, mutual information based on these concepts are discussed. Next, mathematical theorems are proved to demonstrate the validity of the given notions. Thereafter, significance of the features subset is computed by using this mutual information, and corresponding feature selection is suggested to delete the irrelevant and redundant features. The current approach effectively handles noise and subsequent uncertainty in both nominal and mixed data (including both nominal and category variables). Moreover, comprehensive experimental performances are evaluated on real-valued benchmark datasets to demonstrate the practical validation and effectiveness of the addressed technique. Finally, an application of the proposed method is exhibited to improve the prediction of phospholipidosis positive molecules. RF(h2o) produces the most effective results till date based on our proposed methodology with sensitivity, accuracy, specificity, MCC, and AUC of 86.7%, 90.1%, 93.0% , 0.808, and 0.922 respectively.

Feature selection in mixed data: A method using a novel fuzzy rough set-based information entropy

An Emerging Fuzzy Feature Selection Method Using Composite Entropy-Based Uncertainty Measure and Data Distribution

Incremental neighborhood entropy-based feature selection for mixed-type data under the variation of feature set

A Fast Feature Selection Algorithm by Accelerating Computation of Fuzzy Rough Set-Based Information Entropy

Accelerating information entropy-based feature selection using rough set theory with classified nested equivalence classes

U^2F^2S^2 : Uncovering Feature-level Similarities for Unsupervised Feature Selection

Selecting features by utilizing intuitionistic fuzzy Entropy method

Feature Selection with Partition Differentiation Entropy for Large-Scale Data Sets.

Multi-label feature selection based on fuzzy neighborhood rough sets

Feature selection for multiset-valued data based on fuzzy conditional information entropy using iterative model and matrix operation

Feature selection using relative dependency complement mutual information in fitting fuzzy rough set model

Feature Selection with Conditional Mutual Information Considering Feature Interaction

Performance Optimization of a Fuzzy Entropy based Feature Selection and Classification Framework

Hybrid similarity relation based mutual information for feature selection in intuitionistic fuzzy rough framework and its applications

Feature Selection for Unbalanced Distribution Hybrid Data Based on ${K}$-Nearest Neighborhood Rough Set

Feature Selection With Local Density-Based Fuzzy Rough Set Model for Noisy Data

Feature Genes Selection Based on Fuzzy Neighborhood Conditional Entropy

Uncertainty measures and feature selection based on composite entropy for generalized multigranulation fuzzy neighborhood rough set

Interactive and Complementary Feature Selection Via Fuzzy Multigranularity Uncertainty Measures

Feature selection based on multiview entropy measures in multiperspective rough set