An Improved Feature Selection and Classification using Decision Tree for Crop Datasets

Abstract

In this paper a more improved Feature Selection and Classification technique is implanted on Benchmark Datasets such as Mushroom and Soyabean. The Proposed Methodology implemented is based on the Hybrid Combinatorial method of Applying PSO-SVM for the selection of Features from the Dataset and Then Classification is done using Fuzzy Based Decision Tree. Experimental results when performed on Various Datasets prove that the proposed methodology extracts more features as well as provides more accuracy as compared to existing methodologies.