This chapter introduces data mining in agriculture, focusing on support vector machines (SVM). SVM employs optimal hyperplanes in high-dimensional space for linear classification, aiming to find the maximum-margin hyperplane, crucial for accurate classification. The SVM model, a non-probabilistic binary linear classifier, can be adapted for probabilistic classification using methods like Platt scaling. It efficiently handles nonlinear classification through the kernel trick, mapping inputs into high-dimensional spaces. In agriculture, SVM can be used for applications such as crop pest and disease identification, trajectory segmentation, and yield prediction. The chapter underscores SVM's pivotal role in transforming agricultural practices and discusses future research trends.