Aspect-Based Sentiment Analysis (ABSA) for Brand Monitoring: Handling Informal Language in App Review Data

Gandham, Praneeth (2025) Aspect-Based Sentiment Analysis (ABSA) for Brand Monitoring: Handling Informal Language in App Review Data. Masters thesis, Dublin, National College of Ireland.

Preview	PDF (Master of Science) Download (718kB) | Preview
Preview	PDF (Configuration Manual) Download (643kB) | Preview

Abstract

This work presents a hybrid Aspect-Based Sentiment Analysis (ABSA) system that couples a pre-tuned BERT representation with a lexicon-based classifier to analyses informal and noising app review texts. Traditional models typically fail to accommodate slang, abbreviations, and emojis that prevail within user reviews. The proposed hybrid solution addresses such weaknesses with a dedicated preprocessing pipeline and ensemble learning. This experiment with 4,000 labelled app reviews achieved an F1-score of 0.72 from a BERT-based baseline, bettering a lexicon-based baseline with an F1-score of 0.42. The ensemble system further improved, with a macro F1-score of 0.66, and with a overall accuracy rate of 68%. Our experiments confirm our proposed system's effectiveness to accommodate noisiness, aspect-level sentiment. Our work has a number of shortcomings with regards to dataset bias and real-time processing. Our future research includes generalizability across multi-linguals and explanation tools to alleviate bias.

Item Type:	Thesis (Masters)
Supervisors:	Name Email Nagahamulla, Harshani UNSPECIFIED
Uncontrolled Keywords:	Aspect-Based Sentiment Analysis; BERT; Lexicon Model; Hybrid Model; Informal Text; App Reviews; Sentiment Classification; Brand Monitoring
Subjects:	P Language and Literature > P Philology. Linguistics > Computational linguistics. Natural language processing 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 Data Analytics
Depositing User:	Ciara O'Brien
Date Deposited:	25 Aug 2026 13:55
Last Modified:	25 Aug 2026 13:55
URI:	https://norma.ncirl.ie/id/eprint/9630

Actions (login required)

View Item