Detection and Prevention of Negative Comments using Random Forest
Chinni Roshini Durga, Sandeep Vemuri, Veeranki Kavya Lahari · 2024
In the era of social media, the usage of offensive words has increased rapidly during these days. Especially youth play an important role in it. Such kind of comments create a negative impact on the social networks. Cyberbullying also hurts the youth using social media. All these remarks impose a kind of disrespectful culture on social media platforms. Hence, we built a model that detects offensive comments on social media platforms. The System uses Natural Language Processing and Random Forest algorithm to analyze the social comment and classify it as "Hate speech" or "Offensive" or "Normal" comments. It uses an ensemble method - Random Forest for classification of user comments. It classifies whether the comment entered leads to cyberbullying or not, by using the above technologies. This effective classifier acts as the core component in a final prototype system that can detect cyberbullying on social media. It transforms the user comment into a positive form using sentiment analysis and also blocks the user from entering negative comments after some limited attempts. The proposed model converts the negative comment into a positive form without changing the context using sentiment analysis and stores the username for warning the user each time he enters a negative comment. The count of the respective ID that enters a negative comment is also noted. Instead of reporting the user who enters a negative comment, a threshold is set such the user is not allowed to enter a comment beyond that limit.