Multi-Class Sentiment Analysis of Twitter Data Using Bert-Based Model
DOI:
https://doi.org/10.63468/sshrr.266Keywords:
BERT, pre-trained model, sentiment analysis, twitter datasetAbstract
Modern world is a world of social media. X formerly known as Twitter is one of the leading social media platform that represents the public opinions from across the globe on every walk of life. These opinions are not just short texts but represent the sentiments of the public. The sentimental analysis of this large scale text data give insightful information to organizations, researchers, policy makers, and government institutes. Therefore, this work proposed a BERT (Bidirectional Encoder Representations from Transformers)-based sentiment analysis model. The model classifies the X (Twitter) text or tweets as positive, negative or neutral by analyzing the text data. The proposed model is evaluated on accuracy, precision, recall and F1 score. The experiment results showed that the proposed model yielded promising performance against all sentiment classes.
Downloads
Published
Issue
Section
License
Copyright (c) 2025 Shahbaz Hassan Wasti, Ghulam Jillani Ansari, Dr. Jahanzeb Jahan
This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.
All articles published in the Social Sciences & Humanity Research Review (SSHRR) remain the copyright of their respective authors. SSHRR publishes content under the Creative Commons Attribution 4.0 International License (CC BY 4.0), which allows readers to freely share, copy, adapt, and build upon the work in any medium or format, provided proper credit is given to both the authors and the journal.
Third‑party materials included in the articles are subject to their own copyright and must be properly attributed. The journal reserves the right to host, distribute, and preserve all published content to ensure long‑term access and integrity.
