Detecting Offensive Language in Tweets Using Deep Learning

TL;DR

This paper presents an ensemble deep learning approach combining RNN classifiers and user-related features to effectively detect offensive language, such as racism and sexism, in tweets, outperforming existing methods on a large dataset.

Contribution

It introduces a novel ensemble detection scheme that integrates user behavior features with textual analysis for improved offensive language identification.

Findings

Achieves higher classification accuracy than existing methods.

Successfully distinguishes racism and sexism from normal tweets.

Effective on a large, publicly available tweet dataset.

Abstract

This paper addresses the important problem of discerning hateful content in social media. We propose a detection scheme that is an ensemble of Recurrent Neural Network (RNN) classifiers, and it incorporates various features associated with user-related information, such as the users' tendency towards racism or sexism. These data are fed as input to the above classifiers along with the word frequency vectors derived from the textual content. Our approach has been evaluated on a publicly available corpus of 16k tweets, and the results demonstrate its effectiveness in comparison to existing state of the art solutions. More specifically, our scheme can successfully distinguish racism and sexism messages from normal text, and achieve higher classification quality than current state-of-the-art algorithms.