Volume 18, No. 5, 2021
Machine Learning Based Hybrid Ensemble Model Using Majority Voting Technique For Crop Prediction
Purushotam Naidu k , Dr. P Krishna Subba Rao , Dr. MHM Krishna Prasad
Abstract
Agricultural activities employ more than half of India's population. Machine learning algorithms are used in a variety of agricultural studies to predict the best crop. Machine learning is a branch of artificial intelligence that teaches machines to emulate human thought processes. Machine Learning algorithms are often applied to training data to predict outcomes such as crop production, ideal crop to plant, and so on. In this study, a Hybrid Machine Learning Ensemble model for crop prediction is proposed. For crop prediction, we used 22 distinct crops' data, including 7 features such as nitrogen (N), potassium(K), phosphorus(P), temperature, humidity, pH, and rainfall. We proposed a Hybrid Ensemble model with Majority Voting Technique for optimum crop prediction using given dataset. The proposed model is modelled using five different learning algorithms, including Support Vector Machine (SVM), K-Nearest Neighbors (KNN), Naive Bayes, Random Forest (RF), and Logistic Regression, and the prediction accuracy is assessed using the majority vote technique. The prediction accuracy of the hybrid ensemble model was compared to that of seven machine learning methods, including KNN, SVM, Nave Bayes, RF, Decision Tree, Gradient Boosting, and Extreme Gradient Boosting (XGB) classifiers. The results show that the proposed hybrid ensemble model outperformed than other algorithms in terms of accuracy.
Pages: 2282-2293
Keywords: Machine Learning, KNN, SVM, Random Forest, XG Boost, Hybrid ensemble, Majority Voting,