METODE DAN ALGORITMA ANALISIS SENTIMEN: TINJAUAN LITERATUR SISTEMATIS 2023–2026
DOI:
https://doi.org/10.54314/8snx2984Keywords:
analisis sentimen, tinjauan literatur sistematis, pembelajaran mesin, transformer, model bahasa besarAbstract
Abstract: Sentiment analysis has shifted from feature-based machine learning to deep learning, Transformer encoders, and large language models (LLMs), affecting model accuracy and approaches to data formulation, supervision, and evaluation. This systematic literature review maps developments from 2023 to 2026 using an auditable corpus of 30 primary studies classified into four methodological groups: machine learning, deep learning, Transformer, and LLM. The corpus composition changes over time: machine learning accounts for 77.8% of studies in 2023, falls to 40.0% in 2024, and is absent in 2025–2026. Transformer and LLM approaches rise from 0% in 2023 to 50.0% in 2024 and reach 100% in 2025 and 2026. Deep learning serves as a transitional stage, especially in 2023–2024. The evidence indicates a shift from feature engineering and class-balancing optimization toward contextual representations, domain adaptation, prompting, fine-tuning, and assessment of reliability, fairness, and language sensitivity. Greater model capacity, however, does not automatically improve robustness: recent studies highlight prompt sensitivity, demographic differences, and dependence on fine-tuning and domain-specific information. The review recommends controlled comparisons, multilingual evaluation, reproducibility, uncertainty assessment, and resource-aware execution.
Keywords: Sentiment Analysis; Systematic Literature Review; Machine Learning; Transformer; Large Language Models
Abstrak: Perkembangan analisis sentimen bergeser dari pembelajaran mesin berbasis fitur menuju pembelajaran mendalam, enkoder Transformer, dan model bahasa besar (LLM). Perubahan tersebut memengaruhi akurasi model serta pendekatan perumusan data, pembelajaran terawasi, dan evaluasi. Tinjauan literatur sistematis ini memetakan perkembangan pada 2023–2026 berdasarkan 30 artikel penelitian primer yang dapat ditelusuri dan dikelompokkan ke dalam empat kategori metodologis: pembelajaran mesin, pembelajaran mendalam, Transformer, dan LLM. Komposisi kumpulan penelitian berubah dari tahun ke tahun: pembelajaran mesin mencakup 77,8% penelitian pada 2023, turun menjadi 40,0% pada 2024, dan tidak ditemukan pada 2025–2026. Pendekatan Transformer dan LLM meningkat dari 0% pada 2023 menjadi 50,0% pada 2024, kemudian mencapai 100% pada 2025 dan 2026. Pembelajaran mendalam berperan sebagai tahap peralihan, terutama pada 2023–2024. Hasil kajian menunjukkan pergeseran fokus metodologis dari rekayasa fitur dan optimasi keseimbangan kelas menuju representasi kontekstual, adaptasi domain, penyusunan prompt, penyetelan lanjutan (fine-tuning), serta penilaian keandalan, keadilan, dan sensitivitas bahasa. Namun, peningkatan kapasitas model tidak selalu diikuti peningkatan ketahanan. Penelitian terbaru menyoroti sensitivitas terhadap prompt, perbedaan antarkelompok demografis, serta ketergantungan pada penyetelan lanjutan dan informasi khusus domain. Kajian ini merekomendasikan perbandingan terkendali, evaluasi multibahasa, keterulangan penelitian, pengukuran ketidakpastian, dan penerapan yang memperhatikan kebutuhan sumber daya.
Kata kunci: Analisis Sentimen; Tinjauan Literatur Sistematis; Pembelajaran Mesin; Transformer; Model Bahasa Besar.
Downloads
References
Ahmadian, M. S., et al. (2024). Hybrid Models for Emotion Classification and Sentiment Analysis in Indonesian. Applied Computational Intelligence and Soft Computing. https://doi.org/10.1155/2024/2826773.
Al-Khazaleh, M., Alian, M., Biltawi, M., & Al-Hazaimeh, B. (2023). Sentiment Analysis for People’s Opinions about COVID-19 Using LSTM and CNN Models. International Journal of Online and Biomedical Engineering, 19(1), 135–154. https://doi.org/10.3991/ijoe.v19i01.35645.
Ardana, Y., Kambau, R. A., & Mustikasari. (2023). Comparison of Sentiment Analysis of Support Vector Machine and Naïve Bayes Algorithms on Public Response About Online Learning During the COVID-19 Pandemic. AGENTS: Journal of Artificial Intelligence and Data Science, 3(1). https://doi.org/10.24252/jagti.v3i1.46.
Barik, K., Misra, S., Ray, A. K., & Bokolo, A. (2023). LSTM-DGWO-Based Sentiment Analysis Framework for Analyzing Online Customer Reviews. Journal of Environmental and Public Health, 2023, 6348831. https://doi.org/10.1155/2023/6348831.
Darmawan, F., Joe, M., Kurniawan, Y. I., & Afuan, L. (2023). Analisis Sentimen Kemungkinan Depresi dan Kecemasan pada Twitter Menggunakan Support Vector Machine. Jurnal Eksplora Informatika, 13(1), 24–36. https://doi.org/10.30864/eksplora.v13i1.854.
Dewi, T. A., & Mailoa, E. (2023). Perbandingan Implementasi Metode SMOTE pada Algoritma Support Vector Machine (SVM) dalam Analisis Sentimen Opini Masyarakat tentang Mixue. Jurnal Indonesia: Manajemen Informatika dan Komunikasi, 4(3), 849–855. https://doi.org/10.35870/jimik.v4i3.289
Ernawati, N. W., Satya Kumara, I. N., & Setiawan, W. (2023). Perbandingan Metode Klasifikasi Support Vector Machine dan Naïve Bayes pada Analisis Sentimen Kendaraan Listrik. Jurnal SPEKTRUM, 10(3), 106–114. https://doi.org/10.24843/SPEKTRUM.2023.v10.i03.p12.
Febrian, A. T., Purwadi, P., & Kuncoro, A. P. (2026). Comparative Analysis of Data-Level and Cost-Sensitive Learning in IndoBERT-Based Sentiment Analysis of Ruangguru App Reviews. Jambura Journal of Electrical and Electronics Engineering, 8(2), 250–259. https://doi.org/10.37905/jjeee.v8i2.39885
Fehle, J., Kruschwitz, U., Hellwig, N. C., & Wolff, C. (2026). Leveraging Fine-Tuning of Large Language Models for Aspect-Based Sentiment Analysis in Resource-Scarce Environments. Knowledge-Based Systems, 336, 115277. https://doi.org/10.1016/j.knosys.2026.115277
Girard, J. M., Jun, D., Ong, D. C., Liebenthal, E., & Baker, J. T. (2026). Sentiment Analysis of Naturalistic Speech Using Open-Weight Large Language Models. Affective Science, 7(2), 166–180. https://doi.org/10.1007/s42761-025-00352-7.
Hellwig, N. C., Fehle, J., & Wolff, C.
(2024). Exploring Large Language Models for the Generation of Synthetic Training Samples. Expert Systems with Applications, 125514. https://doi.org/10.1016/j.eswa.2024.125514.
Herrera-Poyatos, D., Peláez-González, C., Zuheros, C., Herrera-Poyatos, A., Tejedor, V., Herrera, F., & Montes, R. (2025). An Overview of Model Uncertainty and Variability in LLM-Based Sentiment Analysis: Challenges, Mitigation Strategies and the Role of Explainability. Frontiers in Artificial Intelligence. https://doi.org/10.3389/frai.2025.1609097.
Iskoko, A., Tahyudin, I., & Purwadi, P. (2025a). Hyperparameter Optimization of IndoBERT Using Grid Search, Random Search, and Bayesian Optimization in Sentiment Analysis of E-Government Application Reviews. Jurnal Teknik Informatika (JUTIF), 6(5), 3430–3444. https://doi.org/10.52436/1.jutif.2025.6.5.4897.
Iskoko, A., Tahyudin, I., & Purwadi, P. (2025b). Analisis Sentimen Ulasan Pengguna Aplikasi Sinaga Mobile pada Google Play Store Menggunakan Algoritma Naive Bayes. Jurnal Pendidikan dan Teknologi Indonesia, 5(6), 1635–1645. https://doi.org/10.52436/1.jpti.843.
Kastrati, M., et al. (2026). Do Language Families Matter? Evaluating LLMs for Sentiment Analysis through a Language-Family Lens. Frontiers in Artificial Intelligence. https://doi.org/10.3389/frai.2026.1854873.
Khan, T. A., Sadiq, R., Shahid, Z., Alam, M. M., & Su’ud, M. B. M. (2024).
Sentiment Analysis Using Support Vector Machine and Random Forest. Journal of Informatics and Web Engineering, 3(1). https://doi.org/10.33093/jiwe.2024.3.1.5.
Kusnawi, K., Rahardi, M., & Pandiangan, V. D. (2023). Sentiment Analysis of Neobank Digital Banking Using Support Vector Machine Algorithm in Indonesia. JOIV: International Journal on Informatics Visualization, 7(2), 377–383. https://doi.org/10.30630/joiv.7.2.1652.
Lestari, V. B., Utami, E., & Hanafi. (2024). Combining Bi-LSTM and Word2vec Embedding for Sentiment Analysis Models of Application User Reviews. The Indonesian Journal of Computer Science, 13(1). https://doi.org/10.33022/ijcs.v13i1.3647.
Ningsih, M. R., Unjung, J., Pertiwi, D. A. A., Prasetiyo, B., & Muslim, M. A. (2024). Optimized Support Vector Machine with Particle Swarm Optimization to Improve the Accuracy Amazon Sentiment Analysis Classification. Kinetik, 9(1), 101–108. https://doi.org/10.22219/kinetik.v9i1.1888.
Perwira, R. I., Permadi, V. A., Purnamasari, D. I., & Agusdin, R. P. (2025). Domain-Specific Fine-Tuning of IndoBERT for Aspect-Based Sentiment Analysis in Indonesian Travel User-Generated Content. Journal of Information Systems Engineering and Business Intelligence, 11(1), 30–40. https://doi.org/10.20473/jisebi.11.1.30-40.
Radaideh, M. I., et al. (2025). Fairness and Social Bias Quantification in Large Language Models for Sentiment Analysis. Knowledge-Based Systems, 113569. https://doi.org/10.1016/j.knosys.2025.113569.
Ramadhan, F. E., Karyono, G., & Purwadi, P. (2026). Multilingual Sentiment Detection: IndoBERT Vs. RoBERTa Using NLLB-200 Translations. RABIT: Jurnal Teknologi dan Sistem Informasi Univrab, 11(2), 4855–4865.
Sayarizki, P., Hasmawati, & Nurrahmi, H. (2024). Implementation of IndoBERT for Sentiment Analysis of Indonesian Presidential Candidates. Indonesian Journal on Computing, 9(2), 61–72. https://doi.org/10.34818/indojc.2024.9.2.934.
Sebastian, D. F., Sulistiani, H., & Isnain, A. R. (2024). Sentiment Analysis of Public Opinion on the Right of Inquiry in Indonesia in 2024 Using the Support Vector Machine (SVM) Method. Jurnal Teknik Informatika (JUTIF), 5(4), 1025–1034. https://doi.org/10.52436/1.jutif.2024.5.4.1968.
Simbolon, O. S., Manullang, M. E., Alvarez, S., Brutu, L. F. M., & Indra, E. (2023). Sentiment Analysis of MyPertamina Application Using Support Vector Machine and Naïve Bayes Algorithms. JUSIKOM: Jurnal Sistem Informasi dan Ilmu Komputer, 7(1), 177–189. https://doi.org/10.34012/jurnalsisteminformasidanilmukomputer.v7i1.4078.
Sumantiawan, D. I., Suseno, J. E., & Syafei, W. A. (2023). Sentiment Analysis of Customer Reviews Using Support Vector Machine and SMOTE-Tomek Links for Identify Customer Satisfaction. Jurnal Sistem Informasi Bisnis, 13(1), 1–9. https://doi.org/10.21456/vol13iss1pp1-9.
Tarwoto, Nugroho, Azka, & Graha, W. S. R. (2025). Analisis Sentimen Ulasan Aplikasi Mobile JKN di Google PlayStore Menggunakan IndoBERT. Jurnal JTIK, 9(2), 495–505. https://doi.org/10.35870/jtik.v9i2.3340.
Triawan, P., Tahyudin, I., & Purwadi, P. (2025). Impact of NLP Algorithms on Sentiment Analysis Efficiency and Accuracy. Journal of Information Systems and Informatics, 7(3), 2684–2709. https://doi.org/10.51519/journalisi.v.
Winarto, W., Musdar, I. A., & Hasniati. (2024). Sentiment Analysis of 2024 Presidential Candidate Using the Support Vector Machine Algorithm on Twitter. KHARISMA Tech, 19(1), 86–98. https://doi.org/10.55645/kharismatech.v19i1.464.
Wojdecka, A., Gromadziński, J., & Walczak, K. (2025). Effectiveness of Large Language Models and Software Libraries in Sentiment Analysis. Applied Computer Science,
(2), 128–138. https://doi.org/10.35784/acs_6936.
Zaidan, M. N., Sibaroni, Y., & Prasetyowiyati, S. S. (2024). Learning Rate and Epoch Optimization in the Fine-Tuning Process for IndoBERT’s Performance on Sentiment Analysis of MyTelkomsel App Reviews. Jurnal Teknik Informatika (JUTIF), 5(5), 1443–1450. https://doi.org/10.52436/1.jutif.2024.5.5.2396.
Zhang, W., Deng, Y., Liu, B., Pan, S. J., & Bing, L. (2024). Sentiment Analysis in the Era of Large Language Models: A Reality Check. Findings of NAACL 2024. https://doi.org/10.18653/v1/2024.findings-naacl.246.
Downloads
Published
Issue
Section
License
Copyright (c) 2026 Purwadi Purwadi, Berlilana Berlilana, Bagus Adhi Kusuma, Dhanar Intan Surya Saputra, Hendra Marcos (Author)
This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.
