Multilingual Sentiment Analysis Models using Transfer Learning

Yalla, Rithish Kumar (2024) Multilingual Sentiment Analysis Models using Transfer Learning. Masters thesis, Dublin, National College of Ireland.

Preview	PDF (Master of Science) Download (1MB) | Preview
Preview	PDF (Configuration Manual) Download (1MB) | Preview

Abstract

Sentiment analysis plays an important role in analyzing customer opinions, especially in customer reviews. This work investigates advancements in machine learning models for sentiment classification and introduces an ensemble approach based on transformer-based models, namely XLM-RoBERTa with meta-learning. The proposed transformer-based hybrid model is compared with conventional machine learning algorithms including Naive Bayes, Logistic Regression, and Support Vector Machines on the product reviews dataset. The findings indicate that, while standard models struggle to predict sentiment based on customer review with less than 50% accuracy, the transformer-based hybrid model yields an accuracy of 57% and enhanced classification efficiency in various sentiment classes. While there are some difficulties in correctly categorizing the neutral emotions, the proposed hybrid model shows high performance in identifying the extremely positive and negative emotions. The integration of meta-learning also improves the generality and flexibility of the model. The work also points out the directions for further research including the issues with the class imbalance, the difficulties related to the classification of the neutral sentiments, and the possibilities of real-time learning. It demonstrates the applicability of transformer-based models in sentiment analysis tasks and the basis for further developments of sentiment classification systems for practical use.

Item Type:	Thesis (Masters)
Supervisors:	Name Email Muntean, Cristina Hava UNSPECIFIED
Uncontrolled Keywords:	Natural Language Processing; Hybrid Multilingual Sentiment; XLM-RoBERTa; Meta-Learning
Subjects:	Q Science > QA Mathematics > Electronic computers. Computer science T Technology > T Technology (General) > Information Technology > Electronic computers. Computer science P Language and Literature > P Philology. Linguistics > Computational linguistics. Natural language processing H Social Sciences > HF Commerce > Marketing > Consumer Behaviour Q Science > Q Science (General) > Self-organizing systems. Conscious automata > Machine learning
Divisions:	School of Computing > Master of Science in Data Analytics
Depositing User:	Ciara O'Brien
Date Deposited:	08 Sep 2025 09:15
Last Modified:	08 Sep 2025 09:15
URI:	https://norma.ncirl.ie/id/eprint/8838

Actions (login required)

View Item