Aspect Based Sentiment Analysis using Pre-trained models

Chowdappa Madhugiri, Madhuri (2024) Aspect Based Sentiment Analysis using Pre-trained models. Masters thesis, Dublin, National College of Ireland.

Preview	PDF (Master of Science) Download (1MB) | Preview
Preview	PDF (Configuration Manual) Download (794kB) | Preview

Abstract

Aspect Based Sentiment Analysis is the important task in natural language processing focusing on identifying the sentiments expressed towards specific aspects in the text. This project introduces an advanced ABSA methodology by integrating the pre trained BERT model with Hierarchical Attention Network (HAN) and the Knowledge Graph. The proposed approach enhances the model’s ability to comprehend and analyze context-specific nuances in customer reviews by using the external knowledge and attention mechanisms at both word and sentence levels. The methodology involves the systematic process of data pre-processing, aspect extraction, and sentiment classification and followed by model interpretation using LIME. Experimental results demonstrate that the combined HAN and BERT model significantly improves the sentiment classification accuracy for positive sentiments by offering the detailed insights into the sentiment analysis process. Although the model performs exceptionally well overall there is a slight limitation in detecting negative sentiments which can be addressed in future work. This project contributes to the development of context-aware sentiment analysis models with applications in customer feedback analysis, product reviews and other domains requiring detailed sentiment insights.

Item Type:	Thesis (Masters)
Supervisors:	Name Email Raj, Kislay UNSPECIFIED
Uncontrolled Keywords:	Aspect-Based Sentiment Analysis; Hierarchical Attention Network; Knowledge Graph; BERT; Natural Language Processing; Sentiment Classification; LIME; Data Pre-processing; Model
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 Q Science > Q Science (General) > Self-organizing systems. Conscious automata > Machine learning
Divisions:	School of Computing > Master of Science in Artificial Intelligence
Depositing User:	Ciara O'Brien
Date Deposited:	18 Jun 2025 11:11
Last Modified:	18 Jun 2025 11:11
URI:	https://norma.ncirl.ie/id/eprint/7906

Actions (login required)

View Item