LEVERAGING CONTINUAL FINE-TUNING FOR EMOTION CLASSIFICATION IN PRODUCT REVIEWS ON MSME SUSTAINABILITY SUPPORT

LEVERAGING CONTINUAL FINE-TUNING FOR EMOTION CLASSIFICATION IN PRODUCT REVIEWS ON MSME SUSTAINABILITY SUPPORT

• By Nurohim, Galih Setiawan; Setyadi, Heribertus Ary; Widodo, Pudji; Sutanto, Yusuf;
• Jurnal Nasional Terakreditasi
• 6
• 1382-1390

Analisis otomatis terhadap ulasan produk konsumen sangat penting untuk memahami persepsi
pelanggan yang lebih mendalam melampaui polaritas sentimen dasar. Meskipun model berbasis transformer
telah banyak digunakan dalam analisis sentimen bahasa Indonesia, adaptasinya untuk klasifikasi multiemosi—beralih dari polaritas luas ke status afektif yang spesifik—masih jarang dieksplorasi. Penelitian ini
menjawab celah tersebut dengan mengusulkan pendekatan Continual Fine-Tuning (CFT) untuk mengadaptasi
model IndoBERTweet yang telah dilatih sebelumnya dari tiga kategori sentimen menjadi lima kelas emosi
yang berbeda: Senang (Happiness), Sedih (Sadness), Takut (Fear), Cinta (Love), dan Marah (Anger). Kebaruan
penelitian ini terletak pada pemanfaatan kembali secara strategis (strategic repurposing) bobot model yang
berorientasi sentimen untuk menangkap representasi emosional yang bernuansa dalam diskursus e- commerce di Indonesia. Hasil eksperimen pada dataset PRDECT-ID menunjukkan bahwa model CFT yang
diusulkan mencapai akurasi sebesar 0,8157 dan skor F1 tertimbang sebesar 0,8118, mengungguli jaringan
saraf tradisional dan baseline multibahasa secara signifikan. Meskipun terdapat keterbatasan terkait skala
dataset (5.400 sampel) dan subjektivitas inheren dalam pelabelan emosi, penelitian ini menyediakan kerangka
kerja konseptual yang kuat untuk adaptasi model dalam ekosistem NLP bahasa Indonesia. Temuan ini
menunjukkan bahwa CFT merupakan strategi yang efisien untuk meningkatkan kecerdasan emosional model
transformer, terutama dalam tugas-tugas spesifik domain dengan keterbatasan data berlabel berkualitas
tinggi.

• Katakunci Continual Fine-Tuning, Emotion Classification, Indonesian NLP, IndoBERTweet, Transformer Adaptation.
• URL https://ejournal.nusamandiri.ac.id/index.php/jitk/article/view/7729
• Penerbit JITK Nusa Mandiri, Jakarta Pusat DKI Jakarta

REFERENSI

[1] H. Alqam, M. Razzak, A. Al-Busaidi, and S. AlRiyami, “Conceptualizing Digital Readiness,
Strategic Foresight, and Strategic Flexibility
as Drivers of Digitalization,” Int. J.
Informatics Vis., vol. 8, no. 2, pp. 938 –947,
May 2024, doi: 10.62527/joiv.8.2.2230.
[2] R. Kaur, R. Singh, A. Gehlot, N. Priyadarshi,
and B. Twala, “Marketing Strategies 4.0:
Recent Trends and Technologies in
Marketing,” Sustain., vol. 14, no. 24, pp. 1–17,
2022, doi: 10.3390/su142416356.
[3] A. Romadhony, S. Al Faraby, R. Rismala, U. N.
Wisesti, and A. Arifianto, “Sentiment
Analysis on a Large Indonesian Product
Review Dataset,” J. Inf. Syst. Eng. Bus. Intell. ,
vol. 10, no. 1, pp. 167–178, 2024, doi:
10.20473/jisebi.10.1.167-178.
[4] Y. Wang, E. W. T. Ngai, and K. Li, “The effect
of review content richness on product
review helpfulness: The moderating role of
rating inconsistency,” Electron. Commer. Res.
Appl., vol. 61, p. 4223, 2023, doi:
10.1016/j.elerap.2023.101290.
[5] M. Alzate, M. Arce-Urriza, and J. Cebollada,
“Online reviews and product sales: The role
of review visibility,” J. Theor. Appl. Electron.
Commer. Res., vol. 16, no. 4, pp. 638–669,
2021, doi: 10.3390/jtaer16040038.
[6] L. A. Huwaida et al., “Generation Z and
Indonesian Social Commerce: Unraveling
key drivers of their shopping decisions,” J.
Open Innov. Technol. Mark. Complex. , vol. 10,
no. 2, p. 100256, 2024, doi:
10.1016/j.joitmc.2024.100256.
[7] Y. Mao, Q. Liu, and Y. Zhang, “Sentiment
analysis methods, applications, and
challenges: A systematic literature review,”
J. King Saud Univ. - Comput. Inf. Sci., vol. 36,
no. 4, p. 102048, 2024, doi:
10.1016/j.jksuci.2024.102048.
[8] Q. Zhao, Y. Xia, Y. Long, G. Xu, and J. Wang,
“Leveraging sensory knowledge into Text-
to-Text Transfer Transformer for enhanced
emotion analysis,” Inf. Process. Manag., vol.
62, no. 1, 2025, doi:
10.1016/j.ipm.2024.103876.
[9] E. Latif and X. Zhai, “Fine-tuning ChatGPT for
automatic scoring,” Comput. Educ. Artif.
Intell., vol. 6, no. December 2023, p. 100210,
2024, doi: 10.1016/j.caeai.2024.100210.
[10] N. C. Mei, S. Tiun, and G. Sastria, “Multi-Label
Aspect-Sentiment Classification on
Indonesian Cosmetic Product Reviews with
IndoBERT Model,” Int. J. Adv. Comput. Sci.
Appl., vol. 15, no. 11, pp. 712–720, 2024, doi:
10.14569/IJACSA.2024.0151168.
[11] R. I. Perwira, V. A. Permadi, D. I.
Purnamasari, and R. P. Agusdin, “Domain-
Specific Fine-Tuning of IndoBERT for
Aspect-Based Sentiment Analysis in
Indonesian Travel User-Generated Content,”
J. Inf. Syst. Eng. Bus. Intell. , vol. 11, no. 1, pp.
30–40, 2025, doi: 10.20473/jisebi.11.1.30 -
40.
[12] A. Alamsyah and Y. Sagama, “Empowering
Indonesian internet users: An approach to
counter online toxicity and enhance digital
well-being,” Intell. Syst. with Appl., vol. 22,
no. August 2023, p. 200394, 2024, doi:
10.1016/j.iswa.2024.200394.
[13] F. Baharuddin and M. F. Naufal, “Fine-Tuning
IndoBERT for Indonesian Exam Question
Classification Based on Bloom’s Taxonomy,”
J. Inf. Syst. Eng. Bus. Intell. , vol. 9, no. 2, pp.
253–263, 2023, doi:
10.20473/jisebi.9.2.253 -263.
[14] S. Cahyawijaya et al., “IndoNLG: Benchmark
and Resources for Evaluating Indonesian
Natural Language Generation,” EMNLP 2021
- 2021 Conf. Empir. Methods Nat. Lang.
Process. Proc., pp. 8875–8898, 2021, doi:
10.18653/v1/2021.emnlp-main.699.
[15] Z. Zhou, J. Y. Shin, S. R. Gurudu, M. B. Gotway,
and J. Liang, “Active, continual fine tuning of
convolutional neural networks for reducing
annotation effortsL Active continual fine
tuningcnnrae,” Med. Image Anal., vol. 71, pp.
1–38, 2021, doi:
10.1016/j.media.2021.101997.
[16] K. I. Roumeliotis, N. D. Tselikas, and D. K.
Nasiopoulos, “LLMs in e-commerce: A
comparative analysis of GPT and LLaMA
models in product review evaluation,” Nat.
Lang. Process. J., vol. 6, p. 100056, Mar. 2024,
doi: 10.1016/J.NLP.2024.100056.
[17] D. Aggarwal, S. Damle, N. Goyal, S. Lokam,
and S. Sitaram, “Exploring Continual Fine-
Tuning for Enhancing Language Ability in
Large Language Model,” 2024, [Online].
Available:
https://arxiv.org/abs/2410.16006.
[18] Khamaludin et al., “The influence of social
media marketing, product innovation and
market orientation on indonesian smes
marketing performance,” Int. J. Data Netw.
Sci., vol. 6, no. 1, pp. 9–16, 2021, doi:
10.5267/J.IJDNS.2021.11.002.​