Abstract: As it is known that the backbone the country’s economy is agriculture. Nowadays, the youngsters are not giving importance to the agriculture since the agricultural firms are not fully computerized for enhancing productivity. Hence, the agricultural firms are aimed to computerize their operations in order to increase the productivity. This objective motivates this paper to develop an agriculture classification system with higher accuracy. Moreover, this paper proposes a deferential evolution technique with attribute selection to improve the accuracy in agriculture classification. This proposed system is tested on the various real world benchmark datasets. For evaluating the proposed system the different classification algorithms namely function based logistic classifier, instance classifier namely K-star and tree based logistic model trees (LMT) are used. The experimental results of the proposed system are promising with higher classification accuracy.

Keywords: Agriculture classification system, differential evolution algorithm, agriculture productivity.