Artificial Immune System Algorithms for Crops Classification Using Principal Components Analysis

Abstract

There have been tremendous increase in crop production data which can be used to characterize and predict models in data mining for agriculture. Recently, researchers have shown a lot of interest in applying biologically inspired systems for solving classification and recognition problems. Several solutions have been proposed using Artificial Immune System (AIS), Ant Colony Optimization and so forth in classification problems as another machine learning technique. The field of agriculture is not left behind in the use of machine learning technique for crop and soil classification but few research has been carried out in using AIS as a machine learning technique for crop edibility and disease classification. In this paper, we propose an Artificial Immune System (AIS) solution using AIRS, Clonal and Immnunos algorithms with PCA for crop edibility and crop disease classification. The proposed solution is tested on two crop dataset (Mushroom and Soybeans dataset). The results show significant improvement of the proposed solution over other techniques in most of the cases. Accuracy, true positives and false positives were used as performance measures. The proposed model can be used to enhance crop productivity.