• V.Parameshwar Reddy
• S.Ashokkumar
• B.PanduRangaRaju

Toxic online content has become a major issue intoday’s world due to an exponential increase in the use of internetby people of different cultures and educational background.Differentiating hate speech and offensive language is a keychallenge in automatic detection of toxic text content. In thispaper, we propose an approach to automatically classify tweetson Twitter into three classes: hateful, offensive and clean. UsingTwitter dataset, we perform experiments considering n-gramsas features and passing their term frequency-inverse documentfrequency (TFIDF) values to multiple machine learning modelspropose an approach to devise a machinelearning model which can differentiate between these twoaspects of toxic language. We choose to detect hate speech andoffensive text on Twitter platform. By using publicly availableTwitter datasets we train our classifier model using n-gramand term frequency-inverse document frequency (TFIDF) asfeatures and evaluate it for metric scores

Stress detection, social media, micro-blog,access tokens, and face book.

" A Novel Approach to Detect Hateful Messages in Social Networks", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.5, Issue 12, page no.223-230, December-2018, Available :http://www.jetir.org/papers/JETIR1812534.pdf

" A Novel Approach to Detect Hateful Messages in Social Networks", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.5, Issue 12, page no. pp223-230, December-2018, Available at : http://www.jetir.org/papers/JETIR1812534.pdf