ABSTRACT
This chapter discusses the difference between how the algorithm classifies or predicts outcomes and how the learning that it does on the training set is stored so that it can be applied to new unlabeled data. In machine learning, in so-called parametric based learning, what is learned through training on the training data, is stored in parameters, on the other hand, there are so-called non-parametric algorithms as well, for example KNN. This is an entirely different meaning from the term “parameter” that we have been using for certain elements of the function signature. In the Naive Bayes algorithm, recall that the numerator is a term with many factors. It teaches to split the dataset into two parts such that seventy percent of the dataset will be allocated to the training set and the rest to the test data.
