Perbandingan BERT-base dan DistilBERT untuk Ekstraksi Aspek dan Klasifikasi Sentimen
DOI:
https://doi.org/10.33005/santika.v6i1.1185Keywords:
Sentiment Analysis, Aspect-Based Sentiment Analysis, BERT, DistilBERTAbstract
Aspect-Based Sentiment Analysis (ABSA) merupakan pendekatan analisis sentimen yang bertujuan untuk mengidentifikasi aspek tertentu dari suatu ulasan serta menentukan polaritas sentimen terhadap aspek tersebut. Penelitian ini bertujuan untuk membandingkan kinerja dua model berbasis transformer, yaitu BERT-base dan DistilBERT, dalam tugas Aspect Term Extraction (ATE) dan Aspect Sentiment Classification (ASC) pada dataset ulasan restoran. Dataset yang digunakan terdiri dari total 1006 kalimat, yang kemudian diproses untuk mengekstraksi aspek dan mengklasifikasikan sentimen. Hasil eksperimen menunjukkan bahwa DistilBERT memberikan performa yang sedikit lebih baik dibandingkan BERT-base pada kedua tugas. Pada ATE, DistilBERT memperoleh token F1-score sebesar 0,794, sedangkan BERT-base memperoleh 0,781. Pada ASC, DistilBERT mencapai akurasi sebesar 0,892 dan weighted F1-score sebesar 0,874, lebih tinggi dibandingkan BERT-base dengan akurasi 0,886 dan weighted F1-score 0,856. Selain itu, DistilBERT juga menunjukkan efisiensi waktu pelatihan yang jauh lebih baik dibandingkan BERT-base. Analisis confusion matrix menunjukkan bahwa kedua model lebih akurat dalam memprediksi sentimen positif, sementara performa pada kelas netral masih rendah akibat ketidakseimbangan distribusi data. Hasil penelitian ini menunjukkan bahwa DistilBERT dapat menjadi alternatif yang lebih efisien untuk penerapan ABSA dengan performa yang tetap kompetitif.
References
Jasmine Aulia Mumtaz, Kinaya Khairunnisa Komariansyah, Wildan Holik, Muhammad Galuh Gumelar, Reza Pratama, and Humannisa Rubina Lestari, “Analisis Sentimen Ulasan Aplikasi HeyJapan di Google Play Store Menggunakan Algoritma NLP,” Pragmatik J. Rumpun Ilmu Bhs. dan Pendidikan , vol. 3, no. 3, pp. 157–167, 2025, doi: 10.61132/pragmatik.v3i3.1801.
S. E. Safitri, W. D. Yuniarti, M. R. Handayani, and K. Umam, “User Opinion Mining on the Maxim Application Reviews Using BERT-Base Multilingual Uncased,” J. Sisfokom (Sistem Inf. dan Komputer), vol. 14, no. 3, pp. 365–372, 2025, doi: 10.32736/sisfokom.v14i3.2391.
A. Amrullah, “Advanced Sentiment Analysis Using Deep Learning: A Comprehensive Framework for High-Accuracy and Interpretable Models,” Intellithings, vol. 1, no. 1, p. p, 2025, [Online].
R. Parlika, S. I. Pradika, A. M. Hakim, and K. R. N. M, “Analisis Sentimen Twitter Terhadap Bitcoin dan Cryptocurrency Berbasis Python TextBlob,” vol. 2, pp. 33–37, 2020.
N. Wijaya and E. S. Panjaitan, “Analisis Sentimen Ulasan Aplikasi Instagram di Google Play Store: Pendekatan Multinomial Naive Bayes dan Berbasis Leksikon,” Build. Informatics, Technol. Sci., vol. 6, no. 2, pp. 921–929, 2024, doi: 10.47065/bits.v6i2.5615.
A. N. Hasanah, B. N. Sari, U. S. Karawang, T. Timur, and J. Barat, “Jasa Ojek Online Maxim Pada Google Play,” vol. 12, no. 1, pp. 90–96, 2024.
I. K. Najibulloh, I. Tahyudin, and D. I. S. Saputra, “Analisis Sentimen Ulasan Co-Pilot Google Play dengan SVM, Neural Network, dan Decision Tree,” Edumatic J. Pendidik. Inform., vol. 9, no. 1, pp. 275–283, 2025, doi: 10.29408/edumatic.v9i1.29673.
Ardiansyah, Adika Sri Widagdo, Krisna Nuresa Qodri, F. E. N. Saputro, and Nisrina Akbar Rizky P, “Analisis sentimen terhadap pelayanan Kesehatan berdasarkan ulasan Google Maps menggunakan BERT,” J. Fasilkom, vol. 13, no. 02, pp. 326–333, 2023, doi: 10.37859/jf.v13i02.5170.
A. Karimi, L. Rossi, and A. Prati, “Improving BERT Performance for Aspect-Based Sentiment Analysis,” ICNLSP 2021 - Proc. 4th Int. Conf. Nat. Lang. Speech Process., pp. 39–46, 2021.
J. Devlin, M. W. Chang, K. Lee, and K. Toutanova, “BERT: Pre-training of deep bidirectional transformers for language understanding,” NAACL HLT 2019 - 2019 Conf. North Am. Chapter Assoc. Comput. Linguist. Hum. Lang. Technol. - Proc. Conf., vol. 1, no. Mlm, pp. 4171–4186, 2019.
H. Xu, L. Shu, P. S. Yu, and B. Liu, “Understanding Pre-trained BERT for Aspect-based Sentiment Analysis,” COLING 2020 - 28th Int. Conf. Comput. Linguist. Proc. Conf., pp. 244–250, 2020, doi: 10.18653/v1/2020.coling-main.21.
Y. Wu, Z. Jin, C. Shi, P. Liang, and T. Zhan, “Research on the application of deep learning-based BERT model in sentiment analysis,” Appl. Comput. Eng., vol. 71, no. 1, pp. 14–20, 2024, doi: 10.54254/2755-2721/71/2024ma.
S. Ling, “Fine-Tuning distilBERT for Enhanced Sentiment Classification,” J. Big Data Comput., vol. 2, no. 4, pp. 108–112, 2024, doi: 10.62517/jbdc.202401417.
V. Sanh, L. Debut, J. Chaumond, and T. Wolf, “DistilBERT, a distilled version of BERT: smaller, faster, cheaper and lighter,” pp. 2–6, 2020, [Online]. Available: http://arxiv.org/abs/1910.01108
I. Perikos and A. Diamantopoulos, “Explainable Aspect-Based Sentiment Analysis Using Transformer Models,” Big Data Cogn. Comput., vol. 8, no. 11, 2024, doi: 10.3390/bdcc8110141.
D. Jayakody et al., “Aspect-based Sentiment Analysis Techniques: A Comparative Study,” Moratuwa Eng. Res. Conf. MERCon, pp. 205–210, 2024, doi: 10.1109/MERCon63886.2024.10688631.
J. Serrano-Guerrero, J. A. Olivas, F. P. Romero, and E. Herrera-Viedma, “Sentiment analysis: A review and comparative analysis of web services,” Inf. Sci. (Ny)., vol. 311, pp. 18–38, 2015, doi: 10.1016/j.ins.2015.03.040.
M. Wankhade, A. C. S. Rao, and C. Kulkarni, A survey on sentiment analysis methods, applications, and challenges, vol. 55, no. 7. Springer Netherlands, 2022. doi: 10.1007/s10462-022-10144-1.
M. Zhang et al., “Span-level Aspect-based Sentiment Analysis via Table Filling,” Proc. Annu. Meet. Assoc. Comput. Linguist., vol. 1, pp. 9273–9284, 2023, doi: 10.18653/v1/2023.acl-long.515.
S. K and F. F, “Survey on Aspect-level sentiment analysis,” IEEE Trans. Knowl. Data Eng., vol. 28, no. 3, pp. 813–830, 2016.
Y. C. Hua, P. Denny, J. Wicker, and K. Taskova, A systematic review of aspect-based sentiment analysis: domains, methods, and trends, vol. 57, no. 11. Springer Netherlands, 2024. doi: 10.1007/s10462-024-10906-z.
W. Zhang, X. Li, Y. Deng, L. Bing, and W. Lam, “A Survey on Aspect-Based Sentiment Analysis: Tasks, Methods, and Challenges,” IEEE Trans. Knowl. Data Eng., vol. 35, no. 11, pp. 11019–11038, 2023, doi: 10.1109/TKDE.2022.3230975.
K. Mohiuddin et al., “Attention Is All You Need,” Int. Conf. Inf. Knowl. Manag. Proc., no. Nips, pp. 4752–4758, 2023, doi: 10.1145/3583780.3615497.
D. Bahdanau, K. H. Cho, and Y. Bengio, “Neural machine translation by jointly learning to align and translate,” 3rd Int. Conf. Learn. Represent. ICLR 2015 - Conf. Track Proc., pp. 1–15, 2015.
C. Brun and V. Nikoulina, “Aspect Based Sentiment Analysis into the Wild,” WASSA 2018 - 9th Work. Comput. Approaches to Subj. Sentim. Soc. Media Anal. Proc. Work., pp. 116–122, 2018, doi: 10.18653/v1/P17.
