HBRP Publication Page 1-5 2021. All Rights Reserved Page 1 Journal of Advancement in Software Engineering and Testing Volume 4 Issue 1 Detection and Classification of Online Toxic Comments Omprakash Yadav 1 , Giselle Barretto 2 , Siddhi Bhosle 3* , Candice Dmello 4 1 Professor, 2,3,4 Student Computer Department, Xavier Institute of Engineering, Mumbai, Maharashtra, India. *Corresponding Author E-mail Id:-siddhibhosle99@gmail.com ABSTRACT In the current century, social media has created many job opportunities and has become a unique place for people to freely express their opinions. But as every coin has two sides, the good and the bad, along with the pros, social media has many cons. Among these users, a few bunches of users are taking advantage of this system and are misusing this opportunity to express their toxic mindset (i.e., insulting, verbal sexual harassment, foul behavior, etc.). And hence cyberbullying has become a major problem. If we can filter out the hurtful, toxic words expressed on social media platforms like Twitter, Instagram, and Facebook, the online world will become a safer and more harmonious place. We gained initial ideas by researching current toxic comments classifiers to come up with this design. We then took what we found and made the most user-friendly product possible. For this project, we created a Toxic Comments Classifier which will classify the comments depending on the category of toxicity and will display the percentage of probability for each category of toxicity. Keywords:-Toxic comments classifier, NLP, Word cloud, TF-IDF vectorizer INTRODUCTION Toxic Comments Classifier is a user interface in which an application predicts the probability percentage for each type of toxicity for each comment that is typed by the user within the textbox. In this paper, we will be using Natural Language Processing (NLP) and Flask to deploy our model and solve this problem of identifying the toxicity of online comments. In modern applications, the toxic comments classifier can be a very exceptionally valuable & useful application to eliminate the toxicity and hate from the online world to make it a better place to interact. By providing the percentage-wise probability for each category of toxicity, the aim is to precisely find out the category, the specific comment belongs to and then accordingly eliminate it depending upon the necessity of it. So basically, in our model, there's a textbox within which we type in the comment and click on the predict button. After clicking on the predict button it displays the percentage-wise probability for each category of toxicity depending on the number of toxic words present in the comment. If the majority of the words in the comment are toxic, then the probability will reach up to its maximum limit i.e., 1. If you simply write a straightforward sentence in the textbox, then it will display a low probability percentage of toxicity. STEPS Proposed Work Toxic Comments Classifier uses a Neural Network based algorithm that works on datasets that consists of comments, to administer and eliminate the toxic words from a particular comment by displaying the probability percentage category-wise for a particular comment which is selected