Selecting features by utilizing intuitionistic fuzzy Entropy method

Feature selection is the most significant pre-processing activity, which intends to reduce data dimensionality for enhancing machine learning process. The evaluation of feature must consider classification, performance, efficiency, stability, and many factors. Nowadays, uncertainty commonly occurred in process due time limitations, imprecise information, subjectivity human minds. Moreover, theory intuitionistic fuzzy set has been proven as an extremely valuable tool tackle ambiguity that arises practical situations. Thus, this study introduces a novel framework using entropy. In regard, new entropy IFS proposed first then compared with some previously developed measures. As measure present (features), features higher values are filtered out, remaining having lower have used classify data. To verify effectiveness entropy-based selection, experiments done ten standard benchmark datasets by employing support vector machine, K-nearest neighbor, Naïve Bias classifiers. outcomes validate filter more feasible impressive than existing filter-based methods.