Package: kerntools 1.2.0

Purpose of this document | Introduction | Informal definition | Advantages | Feature space | Kernels for real vectors | Linear kernel | Definition | Usage | RBF kernel | Laplacian kernel | Kernels for real matrices | Frobenius kernel | Kernels for abundances (counts or frequencies) | Bray-Curtis kernel | Ruzicka kernel | Compositional-linear kernel | Aitchison kernel | Kernels for categorical data | Dirac kernel | Kernels for sets | Intersect kernel | Jaccard kernel | Kernels for ordinal data, ranks, and permutations | Kendall's $\tau$ kernel | Kernels for strings, sequences, or (short) texts | Spectrum kernel | Kernels for bag-of-words (BoW) or bags-of-visual-words data | $\chi^2$ kernel

Purpose | Loading | Package Overview | A simple example | A (slightly) more complicated example | Non-standard data, exotic normalizations, and more about feature spaces | Non-standard data | Normalization techniques | Fusing data. A word about a priori and a posteriori feature importances.

Purpose of this document | Introduction | Matrix decomposition | Eigendecomposition | Singular Value Decomposition (SVD) | Principal Components Analysis (PCA) | Kernel PCA | Linear kernel | Rest of kernels | Advantages of kernel PCA | kerntools implementation | Coinertia analysis using kernels | Kernel approach