NLP Steps Import Data Clean Data Vectorizing Split Data into Train and Test Train Model Predict Clean Data Remove Punctuations and replace them with space Remove stopwords like a, the, and, etc. Simplify each word by root of it e.g. liked into like... Its like we are converted into present tense\ Bag of Word Model Creating column of each word RESULTS Support Vector Classifier achieved the highest accuracy of 82.5% while Naive Bayes achieved test accuracy of 68% Important Link: https://towardsdatascience.com/natural-language-processing-nlp-for-machine-learning-d44498845d5b