Abstract

Applicable feature recognizable proof has become a basic assignment to apply data mining calculations viably in true situations. Along these lines, many feature selection methods have been proposed to get the important feature subsets in the writing to accomplish their targets of order and grouping. This paper presents the ideas of feature pertinence, general strategies, assessment standards, and the attributes of feature selection lastly feature selection calculation (utilizing the chi square test ) will be utilized on prediction of school dropouts. The objective of this paper is to discover comparable examples of utilization in the data assembled from datasets and in the end have the option to make predictions for every student dependent on different segment, scholastic and point of view characteristics. In conclusion data from the investigation could reveal insight into how to all the more likely help in danger students. We will finish up this work with genuine application (like early prediction of student dropouts), difficulties, and future research headings of feature selection utilizing filter method.

Description