Performance Analysis of Cuckoo Search Algorithm for Feature Selection of Seeds

Abstract

A reduced number of features are achieved using cuckoo search algorithm. The accuracy has significantly increased even though the number of selected features has drastically reduced. The classification of paddy varieties is done using the selected features. The number of features has been reduced by fifty percent. The main goal was to use a bio-inspired algorithm for feature selection and this is done using the cuckoo search algorithm. The classification accuracy using the selected features achieved is 96.62%.