Method of Profanity Detection Using Word Embedding and LSTM

With the rising number of Internet users, there has been a rapid increase in cyberbullying. Among types cyberbullying, verbal abuse is emerging as most serious problem, for preventing which profanity being identified and blocked. However, users employ words cleverly to avoid blocking. existing discrimination methods, deliberate typos using special characters can be discriminated with high accuracy. they cannot grasp meaning flow sentences, standard such “Sibaljeom (starting point, Korean word that sounds similar swear word)” “Saekkibalgalag (little toe, another are less accurately discriminated. Therefore, order solve this study proposes method discriminating deep learning model context after separating Hangul into onset, nucleus, coda.