Two-stage Bengali sentiment classification: domain adaptation through continual learning and parameter-efficient fine-tuning

Rahman, MD Shaikh and Rabbi, Syed Maudud E and Rashid, Muhammad Mahbubur (2026) Two-stage Bengali sentiment classification: domain adaptation through continual learning and parameter-efficient fine-tuning. In: 2025 20th International Joint Symposium on Artificial Intelligence and Natural Language Processing (iSAI-NLP), 12-14 November 2025, Thailand.

PDF - Published Version Restricted to Repository staff only Download (303kB) | Request a copy

Abstract

Understanding sentiment in low-resource languages remains a key challenge for Natural Language Processing (NLP), particularly when domain-specific data is scarce. In this work, we present SentiBanglaBERT, a two-stage Bengali sentiment classification framework combining domain-adaptive continual pretraining and parameter-efficient fine-tuning. The approach enables contextual adaptation to news-style data while remaining computationally efficient through Low-Rank Adaptation (LoRA). Beyond performance, SentiBanglaBERT integrates SHAP-based interpretability, offering linguistic insights into how Bengali morphological cues—such as negation suffixes and aspectual markers—influence sentiment predictions. Experiments demonstrate stable performance comparable to strong baselines, while providing greater transparency and interpretive depth. This framework highlights the potential of domain-adaptive continual learning as a foundation for interpretable, resource-efficient NLP in morphologically rich, under represented languages. Index Terms—Bengali sentiment analysis, low-resource NLP, domain adaptation continual learning, LoRA, parameter efficient fine-tuning I.

Item Type:	Proceeding Paper (Plenary Papers)
Uncontrolled Keywords:	Bengali sentiment analysis, low-resource NLP, domain adaptation, continual learning, LoRA, parameterefficient fine-tuning
Subjects:	T Technology > TE Highway engineering. Roads and pavements > TE206 Local engineering
Kulliyyahs/Centres/Divisions/Institutes (Can select more than one option. Press CONTROL button):	Kulliyyah of Engineering > Department of Mechatronics Engineering Kulliyyah of Engineering
Depositing User:	Dr Muhammad Rashid
Date Deposited:	04 Feb 2026 11:07
Last Modified:	04 Feb 2026 11:07
Queue Number:	2026-01-Q1976
URI:	http://irep.iium.edu.my/id/eprint/127235

Actions (login required)

View Item