Detecting Hate Speech through Machine Learning

Abstract

Hatred and abusive speeches are identified as huge crime that have been incrementing very recent years, and this has been not only in the specific interaction done face to face but preferably also in the online sharing of information. A considerable number of factors have been contributing to this. There has been a specific study that has well provided a kind of critical overview on the way of detecting such speeches within post or text has been highly evolved over the last few years. However, it has been observed that there are few studies that have been published in detecting hate speech automatically from the perspective of computer science. There is a great way or process with the support of which hate speech can be detected. It has been observed that both the automatic speech recognition and machine learning (ML) have been together complementing each other in the current past, and this has been because both the paradigms are very much deeply ingrained in each other. Therefore, this chapter aims to find out the relationship between hate speech detection and ML and find out the feasible ML algorithms to control hate speeches in social media. It has been observed that there is the implementation of a huge range of several methods of classifying the utilization of the embedding learning for computing all the distances which are semantic in between various parts of the speech which is properly considered to be a specific part of the “othering” narrative. It has also been observed that both the automatic recognition of speech and the ML have been hugely complementing each other in the current past, and this has been just because of the fact that both the paradigms are very much deeply ingrained in each other.