Language-Similarity-Guided Transfer Fine-Tuning of Pre-trained Transformer Models for Sentiment Analysis Across 12 Indonesian Regional Languages
Journal of Computing Theories and Applications · 07.05.2026 · 3 (4) · 547-563
DOI: 10.62411/jcta.15975 ↗Kurz gefasst
Fachartikel in der Zeitschrift Journal of Computing Theories and Applications, online erschienen am 07.05.2026. Verfasst von Brian Rizqi Paradisiaca Darnoto und 1 weiteren Autor:innen. Die Arbeit wurde bisher 4-mal zitiert (Crossref) und nennt 40 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).
Abstract (Originalsprache)
Sentiment analysis for Indonesian regional languages faces two persistent challenges: labeled training data is extremely limited for most regional varieties, and transformer models pre-trained on Bahasa Indonesia do not generalize reliably to languages with substantially different morphological structures. Prior work on the NusaX benchmark has primarily relied on direct fine-tuning, treating each regional language independently and without exploiting linguistic proximity between related languages as a transfer signal. This paper proposes Language-Similarity-Guided Transfer (LSGT), a sequential fine-tuning strategy that first adapts a pre-trained model to a pivot language selected using character trigram similarity, followed by fine-tuning on the target language. Four transformer models are evaluated across all 12 NusaX languages using the official train/validation/test splits: IndoBERT, NusaBERT, mBERT, and XLM-R. Performance is evaluated using four metrics: accuracy, macro F1, macro precision, and macro recall. Experimental results show that LSGT improves macro F1 in 44 of 48 model-language combinations, demonstrating that the fine-tuning strategy itself is a major factor in low-resource cross-lingual sentiment classification. XLM-R benefits most strongly from LSGT, achieving an average improvement of +0.137 macro F1 and a peak gain of +0.298 on Madurese. SHAP-based token attribution analysis further reveals that predictions rely heavily on named entities and domain-specific nouns rather than sentiment-bearing vocabulary, indicating a dataset-level bias inherited from the original SmSA corpus and propagated through the NusaX translation pipeline.
Zitieren
APA
Darnoto, B. R. P., & Firmawan, D. B. (2026). Language-Similarity-Guided Transfer Fine-Tuning of Pre-trained Transformer Models for Sentiment Analysis Across 12 Indonesian Regional Languages. Journal of Computing Theories and Applications, 3(4), 547-563. https://doi.org/10.62411/jcta.15975
Vancouver
Darnoto BRP, Firmawan DB. Language-Similarity-Guided Transfer Fine-Tuning of Pre-trained Transformer Models for Sentiment Analysis Across 12 Indonesian Regional Languages. Journal of Computing Theories and Applications. 2026;3(4):547-563. doi:10.62411/jcta.15975
BibTeX
@article{Darnoto2026,
author = {Darnoto, Brian Rizqi Paradisiaca and Firmawan, Dony Bahtera},
title = {Language-Similarity-Guided Transfer Fine-Tuning of Pre-trained Transformer Models for Sentiment Analysis Across 12 Indonesian Regional Languages},
journal = {Journal of Computing Theories and Applications},
year = {2026},
volume = {3},
number = {4},
pages = {547--563},
doi = {10.62411/jcta.15975}
}
EUVIMED