An Efficient Modeling Based on XGBoost and SVM Algorithms to Predict Crop Yield

Abstract

AbstractIndia is an agricultural country; agriculture is known as the backbone of the Indian economy, where more than 60°/c of the Indian population depend on agriculture for their living. Crop yield prediction is an essential issue in agriculture. Crop yield analysis can be done by using machine learning techniques. Any person in the agriculture field will expect the yield that he is about to get. The things that need to consider for crop yield prediction include analyzing various parameters like pH value to determine soil fertility, location, etc. The number of nutrients is phosphorous (P), nitrogen (N), etc. The soil type, nutrients present in the soil, rainfall, soil composition, and all these things are considered, and all the data mentioned is analyzed. The data is trained by using suitable techniques and creating a model by using machine learning. The proposed system has the best accuracy in crop yield prediction, which shows precise results. The system provides suitable recommendations for the end user about the fertilizers that are fit for good crop yield depending upon the soil’s climate and parameters. This benefits the farmer by increasing and support vector machine (SVM). The support vector machine is a supervised machine learning technique that is used to analyze the data and predict the crop yield in our system. XGBoost algorithm is a machine learning method that refers to “extreme Gradient Boosting.” It tunes the data and gives accurate, efficient, and scalable results. Crop yield production increases the revenue. The system uses two algorithms, known as the XGBoost algorithm.KeywordsSupervised machine learning techniquePredicting crop yieldSupport vector machineXGBoost algorithm introduction