A Comparative Study of Agricultural Crop Yield Prediction Using Machine Learning Techniques

Abstract

Agriculture is the most important factor for survival. Crop yield is a valuable component of food security as the population continues to increase rapidly. Previously, crop and yield predictions were made based on the farmer's experience in a specific location. Farmers will prefer the previous, neighboring, or more trending crop in the surrounding region only for their land, and Farmers are unaware of soil nutrients like nitrogen, phosphorus, and potassium. This can lead to decreased crop growth. So, crop yield prediction is one of agriculture's most pressing issues. Machine learning (ML) is a method that provides a more practical solution to crop yield issues. To boost crop production, agricultural land uses ML algorithms to decide which crops to plant during the growing season. Various machine learning algorithms are used to implement a crop prediction process in this study. Crop yield can be predicted by analyzing various features and Machine Learning (ML) methods like Logistic Regression (LR), Naive Bayes (NB), Support Vector Machine (SVM), Decision Tree (DT), K-Nearest Neighbors Algorithm (KNN), Extreme Gradient Boosting Algorithm (XGBoost) and Random Forest (RF). This work is to show the importance of feature selection and hyperparameter tuning of various ML algorithms. In this study, the Random forest algorithm gives high accuracy when compared with other models.