The impact of social media is a contentious topic, offering both global connectivity and a platform for harmful behavior like cyberbullying and hate speech. In India, with the world’s second-largest internet population, hate speech incidents have increased by 45% in recent years, accompanied by a surge in cybercrimes. These trends have real-life consequences, as seen in the tragic case of an 18-year-old who took his life after receiving hateful comments on social media. Our research paper addresses this gap by conducting a comprehensive comparative analysis of various supervised, unsupervised, and deep learning algorithms for hate speech detection. We aim to find the optimal combination of algorithms that balance accuracy, simplicity, computational efficiency, and ease of implementation. We explored various supervised, unsupervised, and advanced deep learning algorithms to determine the most effective approach for detecting hate speech. This study explored unsupervised learning in this domain, which has received far less attention from other research studies, focusing on three basic models. An extensive comparative analysis has been conducted for the implemented machine learning and deep learning techniques, holding practical importance as this serves as a baseline study to compare different machine learning and deep learning models for automatic hate speech detection in the future. The outcomes of this study are intended to guide researchers and practitioners in developing more robust systems for hate speech detection, contributing to safer online spaces for everyone.