Network-based social media sentimental analysis on twitter

Kotha, Phanisai (2025) Network-based social media sentimental analysis on twitter. Masters thesis, Dublin, National College of Ireland.

Preview	PDF (Master of Science) Download (2MB) | Preview
Preview	PDF (Configuration Manual) Download (443kB) | Preview

Abstract

The purpose of this research is to determine the level of efficiency of the classical machine-learning and the transformer-based models in large-scale Twitter sentiment analysis with the Sentiment140 Dataset. This project aims to address the issues pertaining to the lack of comparative studies focusing on the traditional and transformer models, the growing dependency on social media for analytics, and the understanding of the audience‘s opinion response. A thorough and robust preprocessing application was created, and subsequent to the development of Logistic Regression and the Support Vector Machine, the fine-tuned BERT model was applied. These models were tested and evaluated through metrics of efficiency and accuracy, precision, recall, F1 score, with confusion matrices and ROC/AUC analysis. BERT was proven to be the most accurate with a score of 0.84 and the most outstanding discrimination ability, though the classical models provided a more than competent baseline with a higher degree of efficiency. In order to offer real-time sentiment classification and support the exploration of an interactive knowledge graph for users, a Streamlit application was created. The findings of the research show a strong trade-off between accuracy and efficiency and should help users identify models to improve efficiency based on available resources and performance needs.

Item Type:	Thesis (Masters)
Supervisors:	Name Email Haque, Rejwanul UNSPECIFIED
Uncontrolled Keywords:	Sentiment Analysis; Social Media; Twitter; Natural Language Processing(NLP); Term Frequency–Inverse Document Frequency (TF-IDF); Bidirectional Encoder Representations from Transformers(BERT); Logistic Regression(LR); Support Vector Machine(SVM); Natural Language Toolkit(NLTK); Receiver Operating Characteristic / Area Under the Curve (ROC/AUC)
Subjects:	Q Science > QH Natural history > QH301 Biology > Methods of research. Technique. Experimental biology > Data processing. Bioinformatics > Artificial intelligence Q Science > Q Science (General) > Self-organizing systems. Conscious automata > Artificial intelligence P Language and Literature > P Philology. Linguistics > Computational linguistics. Natural language processing Q Science > Q Science (General) > Self-organizing systems. Conscious automata > Machine learning Z Bibliography. Library Science. Information Resources > ZA Information resources > ZA4150 Computer Network Resources > The Internet > World Wide Web > Websites > Online social networks T Technology > TK Electrical engineering. Electronics. Nuclear engineering > Telecommunications > The Internet > World Wide Web > Websites > Online social networks
Divisions:	School of Computing > Master of Science in Artificial Intelligence
Depositing User:	Ciara O'Brien
Date Deposited:	02 Sep 2026 09:59
Last Modified:	02 Sep 2026 09:59
URI:	https://norma.ncirl.ie/id/eprint/9761

Actions (login required)

View Item