Aspect-Based Sentiment Analysis on Danantara Policies Using IndoBERT–BiLSTM

Alfons Dermawan ER. Laia; Ronsen Purba

doi:10.56741/jnest.v5i03.2102

Aspect-Based Sentiment Analysis on Danantara Policies Using IndoBERT–BiLSTM

Authors

Alfons Dermawan ER. Laia Universitas Mirkoskil
Ronsen Purba Universitas Mirkoskil https://orcid.org/0000-0002-1711-865X

DOI:

https://doi.org/10.56741/jnest.v5i03.2102

Keywords:

ABSA, Deep Learning, LDA, Public Policy, Sentiment Analysis

Abstract

This study examines Aspect-Based Sentiment Analysis (ABSA) of public opinion regarding Danantara's policy on platform X, the primary space for public discourse on strategic economic policies in Indonesia. This research matters because conventional sentiment analysis has not revealed the policy dimensions that attract public attention. The research data consists of 14,123 Indonesian-language tweets from February 24–28, 2025. Aspect identification was conducted exploratively using Latent Dirichlet Allocation (LDA) and focused on three main aspects: policy transparency, management effectiveness, and institutional credibility. Sentiment classification at the aspect level was performed using the IndoBERT–BiLSTM model and compared with the fine-tuned IndoBERT model. The test results showed that IndoBERT–BiLSTM achieved an accuracy of 90.83% and an F1-score of 90.86%, higher than the fine-tuned IndoBERT with an accuracy of 89.51% and an F1-score of 89.43%, resulting in a performance increase of 1.32% and 1.43%. Aspect-by-aspect evaluation showed that IndoBERT–BiLSTM achieved 99% accuracy for policy transparency, 95% for management effectiveness, and 94% for institutional credibility, respectively. These findings confirm the effectiveness of the IndoBERT–BiLSTM integration for the Indonesian-language ABSA and are relevant as a baseline for a more targeted understanding of public perception.

Downloads

Download data is not yet available.

Author Biographies

Alfons Dermawan ER. Laia, Universitas Mirkoskil

is a Master’s candidate in the Information Technology program at Universitas Mikroskil, Medan, Indonesia, where he is researching Aspect-Based Sentiment Analysis (ABSA) of public opinion regarding Danantara's policy on platform X. His research focuses on combining Latent Dirichlet Allocation (LDA) for aspect identification and the IndoBERT–BiLSTM model for sentiment classification, aiming to provide a more nuanced understanding of policy transparency, management effectiveness, and institutional credibility in Indonesian-language social media discourse. (email: alfonsdermawan2002@gmail.com)

Ronsen Purba, Universitas Mirkoskil

is a lecturer in the Master of Information Technology (MTI) Program at Universitas Mikroskil, Medan, Indonesia. His academic interests include information technology, digital systems, data management, and applying technology in education and organizations. He is actively involved in teaching, research, and academic development activities. (email: ronsen@mikroskil.ac.id)

References

S. D. D. Lestari and I. Yuadi, “Mapping sentiment towards Danantara: A combined clustering and text-based predictive model,” J. Law, Pol. Hum., vol. 5, no. 6, pp. 5104–5111, Sep. 2025, doi: 10.38035/jlph.v5i6.2295.

G. G. F. Djema and Ozzi Suria, “Public sentiment analysis of Danantara policy through social media X using SVM and Random Forest,” INOVTEK Polbeng - Seri Informatika, vol. 10, no. 2, pp. 1207–1217, Jul. 2025, doi: 10.35314/rcr21h75.

F. A. D. Aryanti, A. Luthfiarta, and D. A. I. Soeroso, “Aspect-based sentiment analysis with LDA and IndoBERT algorithm on mental health app: Riliv,” J. Appl. Inf. Comput., vol. 9, no. 2, pp. 361–375, Mar. 2025, doi: 10.30871/jaic.v9i2.8958.

D. Nath, S. K. Dwivedi, and A. P. Dwivedi, “Unveiling the impact of Indian government policies using aspect based sentiment analysis with multi-criteria decision-making and hybrid deep learning,” Int. J. Intell. Systems Applications in Eng., vol. 12, no. 22s, pp. 72–82, Jul. 2024, Accessed: Nov. 13, 2025. [Online]. Available: https://ijisae.org/index.php/IJISAE/article/view/6395

R. L. Mustofa, T. L. Mustofa, and C. E. Widodo, “Analisis sentimen berbasis aspek pada aplikasi Elektronik Survei Kepuasan Masyarakat (E-SKM) Jawa Tengah menggunakan Indobert,” J. Teknol. Inf. Ilmu Komputer, vol. 12, no. 4, pp. 867–874, Aug. 2025, doi: 10.25126/jtiik.124.

S. Consoli, L. Barbaglia, and S. Manzan, “Fine-grained, aspect-based sentiment analysis on economic and financial lexicon,” Knowl. Based. Syst., vol. 247, e108781, Jul. 2022, doi: 10.1016/j.knosys.2022.108781.

P. T. Sejati, F. Alzami, A. Marjuni, H. Indrayani, and I. D. Puspitarini, “Aspect-based sentiment analysis for enhanced understanding of ‘Kemenkeu’ tweets,” J. Appl. Inf. Comput., vol. 8, no. 2, pp. 487–498, Nov. 2024, doi: 10.30871/jaic.v8i2.8558.

A. F. Al Farizi and Y. Sibaroni, “Implementation of BiLSTM and IndoBERT for sentiment analysis of TikTok reviews,” JIPI (J. Ilm. Pnl. Pembelajaran Informatika), vol. 10, no. 1, pp. 96–106, Jan. 2025, doi: 10.29100/jipi.v10i1.5815.

H. Ahmadian, T. F. Abidin, H. Riza, and K. Muchtar, “Transformer-based Indonesian language model for emotion classification and sentiment analysis,” in 2023 Int. Conf. on Inf. Technol. Comput. (ICITCOM), IEEE, Dec. 2023, pp. 209–214. doi: 10.1109/ICITCOM60176.2023.10442970.

M. R. Setyawan, F. R. B. Putra, and A. Ramadhani, “Sentiment analysis towards Jokowi post-presidential term using CNN-BiLSTM with multi-head attention on platform X,” ILKOM J. Ilm., vol. 17, no. 2, pp. 150–161, Aug. 2025, doi: 10.33096/ilkom.v17i2.2843.150-161.

O. Alqaryouti, N. Siyam, A. Abdel Monem, and K. Shaalan, “Aspect-based sentiment analysis using smart government review data,” Appl. Comput. Inf., vol. 20, no. 1/2, pp. 142–161, Jan. 2024, doi: 10.1016/j.aci.2019.11.003.

A. A. Chamid, Widowati, and R. Kusumaningrum, “Graph-based semi-supervised deep learning for Indonesian aspect-based sentiment analysis,” Big Data Cognitive Comput., vol. 7, no. 1, e5, Dec. 2022, doi: 10.3390/bdcc7010005.

K. R. A. P. Santoso, A. Husna, N. W. Putri, and N. A. Rakhmawati, “Analisis topik tagar Covidindonesia pada Instagram menggunakan latent dirichlet allocation,” JISKA (J. Informatika Sunan Kalijaga), vol. 7, no. 1, pp. 1–9, Jan. 2022, doi: 10.14421/jiska.2022.7.1.1-9.

L. Zhu, M. Xu, Y. Bao, Y. Xu, and X. Kong, “Deep Learning for Aspect-based sentiment analysis: A review,” PeerJ Comput. Sci., vol. 8, p. e1044, Jul. 2022, doi: 10.7717/peerj-cs.1044.

Y. C. Hua, P. Denny, J. Wicker, and K. Taskova, “A systematic review of aspect-based sentiment analysis: Domains, methods, and trends,” Artif. Intell. Rev., vol. 57, no. 11, e296, Sep. 2024, doi: 10.1007/s10462-024-10906-z.

D. I. Af’idah, P. D. Anggraeni, M. Rizki, A. B. Setiawan, and S. F. Handayani, “Aspect-based sentiment analysis for Indonesian tourist attraction reviews using bidirectional long short-term memory,” JUITA : J. Informatika, vol. 11, no. 1, e27, May 2023, doi: 10.30595/juita.v11i1.15341.

N. Hidayati, “Improving aspect-based sentiment analysis for hotel reviews with latent dirichlet allocation and machine learning algorithms,” Register: J. Ilm. Teknol. Sistem Inf., vol. 9, pp. 144–159, Nov. 2023, doi: 10.26594/register.v9n2.3441.

F. N. Azmina, M. Liebenlito, and D. U. Zulkifli, “Community detection, topic, and sentiment analysis of the Palestine-Israel issue on social media X,” J. Mat., Statis. Komput., vol. 21, no. 2, pp. 502–519, Jan. 2025, doi: 10.20956/j.v21i2.41308.

S. Apriliani, A. Erfina, and C. Warman, “Fine-tuned IndoBERT for aspect-based sentiment analysis of Indonesian five-star hotel reviews,” J. Sisfokom (Sistem Informasi dan Komputer), vol. 14, no. 4, pp. 437–445, Oct. 2025, doi: 10.32736/sisfokom.v14i4.2491.

A. Wafda, D. H. Fudholi, and J. Nugraha, “Aspect-based sentiment analysis on Twitter tweets about the Merdeka Curriculum using IndoBERT,” JITK (J. Ilm. Pengetahuan Teknol. Komput.), vol. 10, no. 3, Feb. 2025, doi: 10.33480/jitk.v10i3.5692.

H. Fei, T.-S. Chua, C. Li, D. Ji, M. Zhang, and Y. Ren, “On the robustness of aspect-based sentiment analysis: Rethinking model, data, and training,” ACM Trans. Inf. Syst., vol. 41, no. 2, pp. 1–32, Apr. 2023, doi: 10.1145/3564281.

A. D. P. Turangan, A. Jacobus, and F. D. Kambey, “Implementation of feature extraction using BERT in aspect based sentiment analysis,” J. EECCIS (Electrics, Electr., Commun., Controls, Inf., Systems), vol. 19, no. 2, pp. 99–104, Aug. 2025, doi: 10.21776/jeeccis.v19i2.1770.

T. Aprilah, D. R. I. M. Setiadi, and W. Herowati, “Enhancing aspect-based sentiment analysis via hugging face fine-tuned IndoBERT,” J. Appl. Inf. Comput., vol. 9, no. 6, pp. 3821–3830, Dec. 2025, doi: 10.30871/jaic.v9i6.11409.

I. Perikos and A. Diamantopoulos, “Explainable aspect-based sentiment analysis using transformer models,” Big Data Cognitive Comput., vol. 8, no. 11, p. 141, Oct. 2024, doi: 10.3390/bdcc8110141.

Hastuti, H. F. Maulana, and E. Satria, “Sentiment analysis and topic modeling of Twitter conversations in Indonesia’s 2024 presidential election,” J. Pekommas, vol. 10, no. 1, Jun. 2025, doi: 10.56873/jpkm.v9i1.5545.

S. Saadah, K, M. Auditama, A. A. Fattahila, F. I. Amorokhman, A. Aditsania, and A. A. Rohmawati, “Implementation of BERT, IndoBERT, and CNN-LSTM in classifying public opinion about COVID-19 vaccine in Indonesia,” J. RESTI (Rekayasa Sistem dan Teknol. Inf.), vol. 6, no. 4, pp. 648–655, Aug. 2022, doi: 10.29207/resti.v6i4.4215.

A. Y. Pratama, G. A. Sanjaya, N. K. Lubis, M. R. Aditya, and Y. Yennimar, “Analisis sentimen publik terkait Danantara menggunakan algoritma IndoBERT pada platform media sosial,” METIK J., vol. 9, 2025. [Online]. Available: https://journal.universitasmulia.ac.id/index.php/metik/article/view/1055

H. Al-Dossari and M. Altalasi, “Multi-aspect sentiment analysis of Arabic Café reviews using machine and deep learning approaches,” Math., vol. 13, no. 24, e3895, Dec. 2025, doi: 10.3390/math13243895.

Md. N. Hoque, R. P. Deb Nath, A. N. Chy, D. Ghose, and M. H. Seddiqui, “Advancing cyberbullying detection in low-resource languages: A transformer-stacking framework for Bengali,” Front. Artif. Intell., vol. 8, Jan. 2026, doi: 10.3389/frai.2025.1679962.