Naive Bayes is a probabilistic algorithm that makes class predictions based on Bayes’ theorem, which states that the probability of a class given a set of features is proportional to the product of the probabilities of each feature given the class. The algorithm assumes independence between the features, so scaling the features should not affect the probabilities that are calculated.
In other words, feature scaling (such as normalization or standardization) changes the units of measurement or distribution of the features, but it does not change the relative ratios of the feature values, which are the primary basis for the calculations in Naive Bayes. The probabilities are based on the frequency of occurrence of feature values in the training data, and scaling the features does not change these frequencies.
So, in summary, the Naive Bayes algorithm is not affected by feature scaling because it is based on probabilities that are not influenced by the scale or distribution of the features.