Analysis of Online Toxicity Detection Using Machine Learning Approaches

TL;DR

This paper reviews machine learning methods for detecting online hate speech, analyzing their trends, shortcomings, and future research directions to improve online toxicity detection.

Contribution

It offers a systematic overview of machine learning approaches for hate speech detection, highlighting research gaps and proposing future directions.

Findings

Identifies key machine learning algorithms used in toxicity detection

Highlights shortcomings in current models and datasets

Suggests future research directions for improved accuracy

Abstract

Social media and the internet have become an integral part of how people spread and consume information. Over a period of time, social media evolved dramatically, and almost half of the population is using social media to express their views and opinions. Online hate speech is one of the drawbacks of social media nowadays, which needs to be controlled. In this paper, we will understand how hate speech originated and what are the consequences of it; Trends of machine-learning algorithms to solve an online hate speech problem. This study contributes by providing a systematic approach to help researchers to identify a new research direction and elucidating the shortcomings of the studies and model, as well as providing future directions to advance the field.