A Comparative Analysis of Tweet Analysis Algorithms Using Natural Language Processing and Machine Learning Models

Online Social Networks (OSNs) have become integral platforms for information sharing, attracting both legitimate users and spammers. Detecting and mitigating spam within these networks pose significant challenges due to the dynamic nature of content and user behavior. In this paper, we present a comprehensive comparative analysis of algorithms for tweet analysis, focusing on Natural Language Processing (NLP) and Machine Learning (ML) models. We evaluate these algorithms through sentiment analysis and multiple attribute analysis, utilizing diverse methodologies and datasets. Our study explores feed-forward neural networks, Bayesian classifiers, and transformer-based models for NLP tasks, achieving high prediction accuracy and insightful metrics such as precision, recall, and F1 score. Furthermore, we delve into multiple attribute analysis using Random Forest, Logistic Regression, and Gradient Boosting algorithms. Through a systematic exploration of various approaches, this work contributes to a deeper understanding of spam detection and sentiment analysis within the context of OSNs, paving the way for enhanced social network security and content analysis.