Paper Key : IRJ************532
Author: Greeshma K S
Date Published: 04 Mar 2024
Abstract
Twitters main objective is to make it possible for everyone to create and share thoughts and data as well as communicate their assumptions and beliefs without restrictions. Hate speech is a text that contains threatening or bad words against a particular group of people in a community.With the growing use of social media, this type of content is growing every day. Hate speech can be harmful to an individual or a community. It is difficult, time-consuming, and error-prone to manually analyze this massive and constantly expanding material. Because of this, it is increasingly important for online social media platforms to screen out any messages that contain hate speech before they are posted to the network.I propose an automated classification system that detects whether a tweet is hateful, offensive, or neither. This approach involves extracting various features like TF-IDF with n gram features, sentiment features, doc2vec features, and enhanced features. These features are used to train machine-learning classifiers. Exhaustive experiments are conducted on the existing Twitter dataset and compared to their accuracy in detecting whether a particular tweet is hateful, offensive, or neither. Also find out the best feature set to perform the same.