Analyzing Bias Trade-Offs in Movie Review Sentiment Analysis using a BERT - SVM Enhanced Model
Jurnal Sistem Cerdas · 19.04.2026 · 9 (1) · 1-13
DOI: 10.37396/jsc.v9i1.570 ↗Kurz gefasst
Fachartikel in der Zeitschrift Jurnal Sistem Cerdas, online erschienen am 19.04.2026. Verfasst von Vany Eka und 1 weiteren Autor:innen.
Abstract (Originalsprache)
Sentiment analysis of movie reviews often exhibits genre-based bias, where model performance varies significantly across subgroups—an issue that standard accuracy metrics can mask. To address this, we propose a novel fairness-aware hybrid model, BERT-SVM (Fairness-Tuned), which integrates sample re-weighting focused on the lowest-performing genre into the BERT-SVM pipeline. Using a public IMDb movie review dataset from Kaggle, we first train a standard BERT-SVM model and identify Horror as the weakest-performing genre (accuracy: 72.3%, vs. overall 89.6%). We then apply targeted re-weighting to upsample underrepresented or misclassified Horror samples during training. The Fairness-Tuned model reduces the accuracy gap by 62%, raising Horror genre accuracy to 83.1% while maintaining strong overall performance (87.4%). This work not only quantifies the fairness–accuracy trade-off but also demonstrates that lightweight, genre-specific bias mitigation within a hybrid architecture can effectively enhance equity without drastic model redesign—highlighting the value of explicit fairness evaluation in NLP applications
Zitieren
APA
Eka, V., & Utama, H. (2026). Analyzing Bias Trade-Offs in Movie Review Sentiment Analysis using a BERT - SVM Enhanced Model. Jurnal Sistem Cerdas, 9(1), 1-13. https://doi.org/10.37396/jsc.v9i1.570
Vancouver
Eka V, Utama H. Analyzing Bias Trade-Offs in Movie Review Sentiment Analysis using a BERT - SVM Enhanced Model. Jurnal Sistem Cerdas. 2026;9(1):1-13. doi:10.37396/jsc.v9i1.570
BibTeX
@article{Eka2026,
author = {Eka, Vany and Utama, Hastari},
title = {Analyzing Bias Trade-Offs in Movie Review Sentiment Analysis using a BERT - SVM Enhanced Model},
journal = {Jurnal Sistem Cerdas},
year = {2026},
volume = {9},
number = {1},
pages = {1--13},
doi = {10.37396/jsc.v9i1.570}
}
EUVIMED