Review on GABASS (Genetic Algorithm Based Attributes Subset Selection) Using Naïve Bayes Classifier

Abstract

ABSTRACT In the present work, review on Genetic Algorithm based Attributes Subset Selection using Naïve Bayes Classifier is presented. A dataset can contain a few components. Many clustering strategies are intended for grouping low-dimensional information. In highdimensional space discovering groups of information objects is trying because of the scourge of dimensionality. At the point when the dimensionality expands, information in the immaterial measurements may create much commotion and veil the genuine bunches to be found. To manage these issues, an effective element subset determination procedure for highdimensional information has been proposed. Feature subset choice decreases the information estimate by evacuating immaterial or repetitive properties. Its point is to enhance the execution consequences of classifiers, however, utilizing an essentially diminished arrangement of components. Genetic algorithm as an optimization tool is proposed to be applied where naïve Bayes classifier will be used to compute the classification accuracy that will be taken as the fitness value of the individual subset. Trials are performed on the bank dataset to group, as indicated by the 11 existing elements. Grouping issues as often as possible have an expansive number of components; however, not every one of them is utile for arrangement. Repetitive and superfluous elements might be lessened the arrangement precision. Feature determination is a procedure of choosing a subset of significant elements, which can diminish the dimensionality, abbreviate the running time, or potentially enhance the order precision.

Keywords: attribute, combination, data sent, information, subset