Developing a classification model to detect hate vs. offensive tweets ready for deployment using various NLP techniques