Benchmarking Model Transformer Modern untuk Analisis Sentimen dan Tren Konsumen dalam Industri Fashion
DOI:
https://doi.org/10.29408/edumatic.v9i3.32657Keywords:
fast fashion, natural language processing, sentiment analysis, transformersAbstract
The dynamic trends and consumer preferences in the fashion industry demand accurate sentiment analysis to support decision-making. This study aims to compare the performance of four Transformer models BERT, DistilBERT, RoBERTa, and DeBERTa, in sentiment classification of fashion product reviews to identify which model best captures the semantic nuances specific to this domain. Using an experimental quantitative approach, we analyzed 278,098 TeePublic reviews that underwent text cleaning, class balancing, and tokenization, and were then split into training, validation, and test sets. The four models were fine-tuned using Optuna-optimized hyperparameters and evaluated with accuracy and F1-score. Our results show that all models achieved nearly identical performance, with F1-scores and accuracy ranging from 0.824 to 0.826. ANOVA testing confirmed no significant differences among models, indicating that increased architectural complexity does not yield substantial benefits for a task of this scale. These findings suggest that performance consistency is driven more by preprocessing quality, class-balancing strategies, and pipeline design. The study offers a comparative basis for model selection in fashion-domain sentiment analysis and highlights that optimizing the processing pipeline often brings more meaningful improvements than adopting larger architectures.
References
Albab, W. U., Mardiah, A. R., Ranjani, G., Karina, G. D., & Safitri, M. N. (2024). Pengaruh Industri Fast Fashion Terhadap Pencemaran Lingkungan dan Penurunan Keadilan Antar Generasi. Indonesian Journal of Criminal Law and Criminology, 5(3), 94–103. https://doi.org/https://doi.org/10.18196/ijclc.v5i3.22830
Aliyu, Y., Sarlan, A., Danyaro, K. U., Sani, A., & Rahman, B. A. (2024). Comparative Analysis of Transformer Models for Sentiment Analysis in Low-Resource Languages. International Journal of Advanced Computer Science and Applications, 15(4). https://doi.org/10.14569/IJACSA.2024.0150437
Ananta, A. Y., Ariyanto, R., Rozi, I. F., & Arianto, R. (2025). Analisis Performa Metode Extreme Learning Machine dan Multiple Linear Regression dalam Prediksi Produksi Gula. Edumatic: Jurnal Pendidikan Informatika, 9(1), 169–178. https://doi.org/10.29408/edumatic.v9i1.29626
Berliana, A. S., & Mustikasari, M. (2024). Analisis Sentimen pada Ulasan Aplikasi JakartaNotebook di Google Play Menggunakan Metode Recurrent Neural Network (RNN). Jurnal Informatika Dan Teknik Elektro Terapan, 12(3), 2830–7062. https://doi.org/10.23960/jitet.v12i3.5067
Braja, A. S. P., & Kodar, A. (2023). Implementasi Fine-Tuning BERT untuk Analisis Sentimen terhadap Review Aplikasi PUBG Mobile di Google Play Store. Jurnal Informatika Merdeka Pasuruan, 7(3), 120. https://doi.org/10.51213/jimp.v7i3.779
Efriadi, D., Rahmaddeni, R., Agustin, A., & Junadhi, J. (2022). Prediksi Penambahan Piutang Iuran Jaminan Sosial Ketenagakerjaan menggunakan Algoritma K-Nearest Neighbor. Edumatic: Jurnal Pendidikan Informatika, 6(1), 49–57. https://doi.org/10.29408/edumatic.v6i1.5255
Hakim, A. L., & Rusadi, E. Y. (2022). Kritik Globalisasi: Fenomena Fast Fashion Sebagai Budaya Konsumerisme Pada Kalangan Pemuda Kota Surabaya. ALMAARIEF, 59-67. https://doi.org/10.35905/almaarief.v4i2.2768
Joshy, A., & Sundar, S. (2022). Analyzing the Performance of Sentiment Analysis using BERT, DistilBERT, and RoBERTa. International Power and Renewable Energy Conference (IPRECON), 1-6. India: IEEE https://doi.org/10.1109/IPRECON55716.2022.10059542
Jumansyah, L., Mohamad Soleh, A., Dyah Syafitri, U., Risman Dwi, L., Mohamad, A., & Dyah, U. (2024). Manifold Learning and Undersampling Approaches for Imbalanced Class Sentiment Classification. Knowledge Engineering and Data Science, 7(2), 139–151.
Kadek, N., & Diantari, Y. (2021). Trend Cycle Analysis on Fast Fashion Products. Journal of Aesthetics, Design, and Art Management, 1(1), 24–33. https://doi.org/10.58982/jadam.v1i1.101
Karimah, N. (2024). Multi-Aspect Sentiment Analysis Pada Review Film Menggunakan Metode Bidirectional Encoder Representations from Transformers (BERT). Komputika : Jurnal Sistem Komputer, 13(1), 63–72. https://doi.org/10.34010/komputika.v13i1.11098
Khadapi, M., & Pakpahan, V. M. (2024). Analisis Sentimen Berbasis Jaringan LSTM dan BERT terhadap Diskusi Twitter tentang Pemilu 2024. Jurnal Komputer dan Informatika. Jurnal Komputer dan Informatika, 6(2), 130–137.
Kim, H. J., Cho, H., Lee, S. W., Kim, J., Park, C., Lee, S. G., Yoo, K. M., & Kim, T. (2023). Universal Domain Adaptation for Robust Handling of Distributional Shifts in NLP. Findings of the Association for Computational Linguistics: Empirical Methods in Natural Language Processing 2023, 1(1), 5888–5905. https://doi.org/10.18653/v1/2023.findings-emnlp.392
Kornelis, Y. (2022). Fenomena Industri Fast Fashion: Kajian Hukum Perspektif Kekayaan Intelektual Indonesia. Jurnal Komunitas Yustisia, 5(1), 262–277. https://doi.org/10.23887/jatayu.v5i1.46040
Kumar, D., & Weissenberger-Eibl, M. (2024, June 19). Artificial Intelligence Driven Trend Forecasting: Integrating BERT Topic Modelling and Generative Artificial Intelligence for Semantic Insights. R&D Management Conference 2024. Stockholm: Fraunhover. https://doi.org/10.24406/publica-3456
Majid, A., Nugraha, D., & Adhinata, F. D. (2023). Sentiment Analysis on Tiktok Application Reviews Using Natural Language Processing Approach. Journal of Embedded Systems, Security and Intelligent Systems, 4(1), 32–38. https://doi.org/10.59562/jessi.v4i1.471
Mangal, D., & Makwana, H. (2025). Performance Analysis of Different BERT Implementation for Event Burst Detection from Social Media Text. Indonesian Journal of Electrical Engineering and Computer Science, 38(1), 439–446. https://doi.org/10.11591/ijeecs.v38.i1.pp439-446
Nabila, P., Wajidi, F., & Firgiawan, W. (2025). Analisis Algoritma K-Nearest Neighbor dan Naive Bayes pada Kebijakan Perkuliahan Online dengan Multi Bahasa. Techno.Com, 24(2), 524–542. https://doi.org/10.62411/tc.v24i2.12656
Pangestu, A. G., Winarno, S., Nugraha, A., & Muttaqin, A. N. I. (2025). DiabTrack: Sistem Prediksi Dini Diabetes Melitus Tipe 2 berbasis Web menggunakan Algoritma K-Nearest Neighbors. Edumatic: Jurnal Pendidikan Informatika, 9(1), 284–293. https://doi.org/10.29408/edumatic.v9i1.29691
Rahayu, S., MZ, Y., Bororing, J. E., & Hadiyat, R. (2022). Implementasi Metode K-Nearest Neighbor (K-NN) untuk Analisis Sentimen Kepuasan Pengguna Aplikasi Teknologi Finansial FLIP. Edumatic: Jurnal Pendidikan Informatika, 6(1), 98–106. https://doi.org/10.29408/edumatic.v6i1.5433
Ramadhan, A. F. (2024). Tren Fast Fashion Pakaian Masa New Normal di Indonesia: Efektivitas Konsep Sustainable Fashion Terhadap Lingkungan. Journal of Waste and Sustainable Consumption, 1(2), 77–89. https://doi.org/10.61511/jwsc.v1i2.2024.1247
Riaz, N., Mehboob, H., & Wajid, S. (2024). Natural Language Processing For Sentiment Analysis: A Machine Learning Approach. Contemporary Journal of Social Science Review, 2(04), 106-116.
Rudniy, A., Rudna, O., & Park, A. (2024). Trend tracking tools for the fashion industry: the impact of social media. Journal of Fashion Marketing and Management, 28(3), 503–524. https://doi.org/10.1108/JFMM-08-2023-0215
Rufikasari, Y. D. (2022). Telaah Teologi, Ekonomi dan Ekologi Terhadap Fenomena Fast Fashion Industry. Teologis, Relevan, Aplikatif, Cendikia, Kontekstual, 1(2), 64–83. https://doi.org/10.61660/tep.v1i2.23
Soni, S., & Baldawa, S. (2023). Analyzing Sustainable Practices in Fashion Supply Chain. International Research Journal of Business Studies, 16(1), 11–25. https://doi.org/10.21632/irjbs.16.1.11-25
Wang, D., & Chen, G. (2025). Evaluating the Use of BERT and Llama to Analyse Classroom Dialogue for Teachers’ Learning of Dialogic Pedagogy. British Journal of Educational Technology, 56(6), 2671–2704. https://doi.org/10.1111/bjet.13604
Downloads
Published
How to Cite
Issue
Section
License
Copyright (c) 2025 Daniel Fernando Herawan, Theresia Ratih Dewi Saputri
This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.
All articles in this journal are the sole responsibility of the authors. Edumatic: Jurnal Pendidikan Informatika can be accessed free of charge, in accordance with the Creative Commons license used.
This work is licensed under a Lisensi a Creative Commons Attribution-ShareAlike 4.0 International License.
