Class of algorithms for pattern analysis (eg. SVM). **Kernel trick**: transforming data into another dimension that has a clear dividing margin between classes of data, without computing the coordinates of the data in that space, but the inner products between the images of all pairs of data in the feature space (using a user-defined similarity function, the "kernel function")
Kernel methods are powerful learning methodologies that provide **a simple way to construct nonlinear algorithms from linear ones**. Despite their popularity, they suffer from **poor scalability in big data scenarios** ([src](https://arxiv.org/abs/1706.06296)).
**Kernel trick**: Kernel functions enable to operate in a high-dimensional, implicit feature space without computing the coordinates of the data in that space, by simply computing the inner products between the images of all pairs of data in the feature space.
Algorithms capable of operating with kernels include SVM, Gaussian processes,PCA, spectral clustering... Any linear model can be turned into a non-linear model by applying the kernel trick to the model: replacing its features (predictors) by a kernel function.