kernlab is an extensible R package for kernel-based machine learning methods. It leverages R's S4 object system to provide a framework for creating and using kernel-based algorithms. The package includes implementations of various kernel functions, support vector machines (SVM), relevance vector machines, Gaussian processes, ranking algorithms, kernel PCA, kernel CCA, kernel feature analysis, online kernel methods, and spectral clustering. It also offers a general-purpose quadratic programming solver and an incomplete Cholesky decomposition method. The package supports both standard and user-defined kernels, making it versatile for a wide range of applications. Key features include efficient kernel matrix computation, kernel utility methods, and a variety of kernel-based methods for classification, regression, and clustering.kernlab is an extensible R package for kernel-based machine learning methods. It leverages R's S4 object system to provide a framework for creating and using kernel-based algorithms. The package includes implementations of various kernel functions, support vector machines (SVM), relevance vector machines, Gaussian processes, ranking algorithms, kernel PCA, kernel CCA, kernel feature analysis, online kernel methods, and spectral clustering. It also offers a general-purpose quadratic programming solver and an incomplete Cholesky decomposition method. The package supports both standard and user-defined kernels, making it versatile for a wide range of applications. Key features include efficient kernel matrix computation, kernel utility methods, and a variety of kernel-based methods for classification, regression, and clustering.