This paper addresses the analysis of hate speech detection models, responding to their increasing presence on social media and the need for effective countermeasures. While various techniques exist for automated hate speech detection, aimed at classifying text into hate speech or non-hate speech and identifying related attributes, a significant performance gap persists between these categories. The inherent challenge lies in the scarcity of distinct hate speech features, relegating it to a minority within datasets. To mitigate this, we propose employing Decision Trees, K-nearest neighbors (KNN), and Random Forest classification models. Our evaluation encompasses these models’ performance on extensive hate speech datasets sourced from common texts. Leveraging Python libraries such as Word Cloud, Sklearn, Pandas, Seaborn, Numpy, Lazypredict, Confusion Matrix, and F1 Score, we assess the realtime predictive accuracy of the hate speech detection model.