Algorithm has deep learning network(Feet forward neural network).
Tokenization- Splitting a string into meaningful units. (Punctuation characters works different for every tokenization technique)
Stemming(Porter Stemer Algorithm)- A NLP techniques, generates the root form of the words. Crude heuristic that chops of the ends off of words.
PreProcessing Pipeline--- Tokenize->Lower+Stem->Excluding Punctuation characters->Bag of words