Analysis of Real Time Twitter Sentiments using Deep Learning Models

Abstract

Understanding attitudes regarding distinct topics and public opinions on the sentimental analysis of social media data is important. This research analyses the real-time twitter sentiments using deep learning. The major objective of the study is to create an efficient sentiment analysis algorithm to accurately ensure the sentiment polarity (positive, neutral or negative) of tweets. This study proposed a deep learning approach to capture the contextual information and complex patterns in social media data which leverages the power of neutral networks. To assess the performance of the algorithm the study relies on the evaluation of F1 score, accuracy, precision, and recall through rigorous evaluation metrics. The efficiency of the proposed approach is demonstrated by the numerical outcomes of the study. A novel contribution is provided with a specific emphasis on real-time Twitter sentiments by the study to enhance the sentiment analysis techniques for social media data. The significant implication from accurate and timely analysis of Twitter sentiments for several applications includes public opinion tracking, brand management, customer feedback analysis, and reputation monitoring. The potential to provide significant insights to researchers, organisations and business can be made from promptly addressing the sentiments expressed on real time data of twitter.