An Ensemble model to analyze the sentiments of Textual Data for Monitoring and Comprehending Racist Views

Sentiment analysis, which seeks to examine and comprehend the underlying sentiment or emotion portrayed in textual data, has grown to be a significant topic of research in natural language processing. The proliferation of user-generated information on social media sites like Twitter/X makes them an ideal source for sentiment analysis investigations. This abstract provides a thorough analysis of sentiment analysis using a Twitter/X dataset with a comparison of the effectiveness of conventional machine learning models as the main topic. This study's main goal is to categorize posts as racist or sexist or non-racist or non-sexist. The Twitter/X dataset is preprocessed to eliminate noise such as stop words, URLs, mentions, and hashtags before being tokenized to extract useful features. Evaluation criteria like accuracy, F-score are used to evaluate each model's performance. The experimental findings and comparison analyses offer insightful information for real-world uses like social media sentiment analysis.