OPINIONNET: MACHINE LEARNING APPROACHES TO TWITTER SENTIMENT EXPLORATION

Abstract

Social media platforms like Twitter generate massive volumes of user-generated content reflecting public opinion, sentiments, and trends. Understanding these sentiments is crucial for businesses, policymakers, and researchers to gauge public reaction and make informed decisions. This study introduces OpinionNet, a machine learning–based framework designed to analyze and classify sentiments expressed in tweets. The system leverages natural language processing (NLP) techniques for preprocessing, feature extraction, and text representation, combined with supervised machine learning models such as Support Vector Machines, Random Forests, and Neural Networks for sentiment classification. Experimental evaluations on large-scale Twitter datasets demonstrate that OpinionNet achieves high accuracy, precision, and recall in distinguishing positive, negative, and neutral sentiments. The findings highlight the framework’s effectiveness in extracting actionable insights from social media data, enabling organizations to monitor trends, detect emerging issues, and understand public opinion in real time.