Authors : Sharmas Vali K.; Dr. Girish Kumar D.

Volume/Issue : Volume 11 - 2026, Issue 5 - May

Google Scholar : https://tinyurl.com/2zue9325

Scribd : https://tinyurl.com/4t7makud

DOI : https://doi.org/10.38124/ijisrt/26May648

Note : A published paper may take 4-5 working days from the publication date to appear in PlumX Metrics, Semantic Scholar, and ResearchGate.

Abstract : The creation of text data that reflects various public thoughts and feelings has been steadily increasing in recent years due to the development and expansion of various social media platforms. Because of its unstructured form and speed, it is challenging to analyze such data in real-time and get valuable insights from it. Traditional sentiment analysis methods are typically offline-based, which makes them inappropriate and ineffective in these situations. This study proposes a Real-Time Sentiment A monitoring system that employs a number of natural language processing methods to process and update real-time text input and uses various transformer models to classify it as positive, negative, or neutral. The suggested method is effective and appropriate for real-time applications, according to the experimental findings. Index Terms: Transformer Models, Natural language processing, sentiment analysis, Real-Time Systems, and Social Media Analytics.

Keywords : Sentiment Analysis, Natural Language Processing, Real-Time Systems, Social Media Analytics, Transformer Models.

References :

1. B. Pang and L. Lee, “Opinion Mining and Emotion Analysis,” Informa-tion Foundations and Trends Retrieval, 2008.
2. B. Liu, “Sentiment Analysis and Opinion Mining,” Morgan and Clay-pool, 2012.
3. J. Devlin et al., “BERT:Deep Bidirectional Transformers for Language Understanding: Pre-training, 2019.
4. A. Radford et al., “Language Models are Unsupervised Multitask Learners,” 2019.