NLP API to return probability that if text is toxic, obscene, insulting or threatening
Combatting toxicity online.