Modified BERTopic using IndoSBERT for topic modeling in Bahasa

Authors

  • Nur Huda Riyantoni UIN Sunan Ampel Surabaya, Indonesia
  • Khalid Sjamsuri UIN Sunan Ampel Surabaya, Indonesia
  • Subhan Nooriansyah UIN Sunan Ampel Surabaya, Indonesia

DOI:

https://doi.org/10.21107/kursor.v13i3.497

Keywords:

Bahasa Indonesia, BERTopic, IndoSBERT, Information System, Topic Modelling

Abstract

The vast amount of textual data -particularly undergraduate theses abstracts produced in the digital age- makes it difficult for readers to identify the topics contained within them. Topic modeling facilitates readers in identifying topics within a collection of textual data. One method for topic modeling is BERTopic. BERTopic is a framework for topic modeling that utilizes the BERT model in embedding stage. This study use IndoSBERT and multilingual SBERT in the BERTopic embedding stage to determine which model performs better in generating topic for a dataset of Indonesian-language undergraduate theses abstract. The topic generated using these IndoSBERT and multilingual SBERT embedding are then evaluated using the topic coherence and topic diversity metrics. The research results show that topics generated by IndoSBERT have higher topic coherence and topic diversity scores, than those generated by multilingual SBERT. These results indicate that IndoSBERT is better in generating topics with topic coherence and topic diversity than multilingual SBERT. The contribution of this research lies in the modification of the BERTopic embedding stage using IndoSBERT for topic modelling in Bahasa. This is because the use of IndoSBERT has so far been limited for text classification task.
Key words: Bahasa Indonesia, BERTopic, IndoSBERT, Information System, Topic modelling.

Downloads

Download data is not yet available.

References

[1] Rosidah, “Abstrak Dalam Skripsi Mahasiswa Universitas Lampung Tahun 2015 Dan Implikasinya Terhadap Mata Kuliah Umum Bahasa Indonesia Di Perguruan Tinggi,” Skripsi, FAKULTAS KEGURUAN DAN ILMU PENDIDIKAN, UNIVERSITAS LAMPUNG, 2016. Accessed: Mar. 26, 2026. [Online]. Available: https://digilib.unila.ac.id/23679/

[2] Ach. Khozaimi, Y. D. Pramudita, and F. Solihin, “DESIGN AND DEVELOPMENT OF BACKEND APPLICATION FOR THESIS MANAGEMENT SYSTEM USING MICROSERVICE ARCHITECTURE AND RESTFUL API,” J. Ilm. Kursor, vol. 11, no. 4, pp. 179–186, Jan. 2023, doi: 10.21107/kursor.v11i4.313.

[3] O. Babalola, B. Ojokoh, and O. Boyinbode, “Comprehensive Evaluation of LDA, NMF, and BERTopic’s Performance on News Headline Topic Modeling,” J. Comput. Theor. Appl., vol. 2, no. 2, pp. 268–289, Nov. 2024, doi: 10.62411/jcta.11635.

[4] R. Muthusami, N. Mani Kandan, K. Saritha, B. Narenthiran, N. Nagaprasad, and K. Ramaswamy, “Investigating topic modeling techniques through evaluation of topics discovered in short texts data across diverse domains,” Sci. Rep., vol. 14, no. 1, p. 12003, May 2024, doi: 10.1038/s41598-024-61738-4.

[5] M. M. Mostafa, M. Alhur, and A. M. Moustafa, “A structural topic modeling of communication research: insights from over a century of journals’ abstracts,” Int. J. Inf. Manag. Data Insights, vol. 5, no. 2, p. 100364, Dec. 2025, doi: 10.1016/j.jjimei.2025.100364.

[6] D. L. M. Owa, “Identification of Topics from Scientific Papers through Topic Modeling,” Open J. Appl. Sci., vol. 10, no. 04, pp. 541–548, 2021, doi: 10.4236/ojapps.2021.104038.

[7] Gaziantep Üniversitesi, K. Gökdağ, Gaziantepunıversıty - Faculty of Educatıon, and M. F. Özmantar, “Emerging Research Themes in Mathematics Education: A Topic Modeling Analysis of Most Influential Journals (2019-2023),” Int. J. Progress. Educ., vol. 20, no. 6, pp. 16–32, Dec. 2024, doi: 10.29329/ijpe.2024.1078.2.

[8] R. M. Suleman and I. Korkontzelos, “Extending latent semantic analysis to manage its syntactic blindness,” Expert Syst. Appl., vol. 165, p. 114130, Mar. 2021, doi: 10.1016/j.eswa.2020.114130.

[9] A. S. Nugroho, A. Saikhu, and R. N. E. Anggraini, “Development of Reviewer Assignment Method with Latent Dirichlet Allocation and Link Prediction to Avoid Conflict of Interest,” J. RESTI Rekayasa Sist. Dan Teknol. Inf., vol. 7, no. 4, pp. 837–844, Aug. 2023, doi: 10.29207/resti.v7i4.4900.

[10] S. Bellaouar, M. M. Bellaouar, and I. E. Ghada, “Topic Modeling: Comparison of LSA and LDA on Scientific Publications,” in 2021 4th International Conference on Data Storage and Data Engineering, Barcelona Spain: ACM, Feb. 2021, pp. 59–64. doi: 10.1145/3456146.3456156.

[11] M. D. R Wahyudi, A. Fatwanto, U. Kiftiyani, and M. Galih Wonoseto, “Topic Modeling of Online Media News Titles during COVID-19 Emergency Response in Indonesia Using the Latent Dirichlet Allocation (LDA) Algorithm,” Telematika, vol. 14, no. 2, pp. 101–111, Aug. 2021, doi: 10.35671/telematika.v14i2.1225.

[12] S. A. Nugroho, F. A. Bachtiar, and R. C. Wihandika, “ASPECT EXTRACTION IN E-COMMERCE USING LATENT DIRICHLET ALLOCATION (LDA) WITH TERM FREQUENCY-INVERSE DOCUMENT FREQUENCY (TF-IDF),” J. Ilm. Kursor, vol. 11, no. 2, p. 53, Jan. 2022, doi: 10.21107/kursor.v11i2.247.

[13] M. Grootendorst, “BERTopic: Neural topic modeling with a class-based TF-IDF procedure,” 2022, doi: 10.48550/ARXIV.2203.05794.

[14] M. de Groot, M. Aliannejadi, and M. R. Haas, “Experiments on Generalizability of BERTopic on Multi-Domain Short Text,” Dec. 16, 2022, arXiv: arXiv:2212.08459. doi: 10.48550/arXiv.2212.08459.

[15] A. F. A. Mahmoud et al., “A Comparative Evaluation of LDA, NMF, and BERTopic: Analyzing Perplexity and Coherence Metrics,” Ingénierie Systèmes Inf., vol. 30, no. 12, pp. 3163–3169, Dec. 2025, doi: 10.18280/isi.301208.

[16] B. Ogunleye, T. Maswera, L. Hirsch, J. Gaudoin, and T. Brunsdon, “Comparison of Topic Modelling Approaches in the Banking Context,” Appl. Sci., vol. 13, no. 2, Art. no. 2, Jan. 2023, doi: 10.3390/app13020797.

[17] R. Egger and J. Yu, “A Topic Modeling Comparison Between LDA, NMF, Top2Vec, and BERTopic to Demystify Twitter Posts,” Front. Sociol., vol. 7, p. 886498, May 2022, doi: 10.3389/fsoc.2022.886498.

[18] A. Krishnan, “Exploring the Power of Topic Modeling Techniques in Analyzing Customer Reviews: A Comparative Analysis,” 2023, arXiv. doi: 10.48550/ARXIV.2308.11520.

[19] J. Devlin, M.-W. Chang, K. Lee, and K. Toutanova, “BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding,” 2018, doi: 10.48550/ARXIV.1810.04805.

[20] T. Mastery, “Bertopic – A Must Read Comprehensive Guide,” DotCom magazine. [Online]. Available: https://dotcommagazine.com/2023/07/bertopic-a-must-read-comprehensive-guide/

[21] I. F. Rozi, I. Maulidia, M. Hani’ah, R. Arianto, D. R. Yunianto, and A. Y. Ananta, “Comparison of Feature Extraction in Support Vector Machine (SVM) Based Sentiment Analysis System,” J. Ilm. Kursor, vol. 13, no. 1, pp. 1–12, Jul. 2025, doi: 10.21107/kursor.v13i1.417.

[22] A. Z. A. Muhabbab and R. A. F. Rizki, “Topic Modelling Berbasis Embedding pada Komentar YouTube,” Semin. Nas. Off. Stat., vol. 2024, no. 1, pp. 873–884, Nov. 2024, doi: 10.34123/semnasoffstat.v2024i1.2286.

[23] W. M. Sihombing and T. Widiyaningtyas, “Sentiment Analysis and Topic Modelling Using IndoBERTweet and BERTopic for Public Health Issues: Analisis Sentimen dan Pemodelan Topik Menggunakan IndoBERTweet dan BERTopic untuk Isu Kesehatan Publik,” Indones. J. Innov. Stud., vol. 26, no. 4, Nov. 2025, doi: 10.21070/ijins.v26i4.1833.

[24] D. Aryani, I. Lucia Kharisma, A. Sujjada, and K. Kamdan, “Topic Modeling of the 2024 Election Using the BERTopic Method on Detik.com News Articles,” Inf. J. Ilm. Bid. Teknol. Inf. Dan Komun., vol. 9, no. 2, pp. 171–180, Aug. 2024, doi: 10.25139/inform.v9i2.8429.

[25] W. Wahyuni, T. P. Lestari, M. Apriliana, and R. Gumelta, “Implementation of BERTopic for Topic Modeling Analysis of the Free Nutritious Meal Program Based on YouTube Comments,” J. Appl. Inform. Comput., vol. 9, no. 4, pp. 1964–1971, Aug. 2025, doi: 10.30871/jaic.v9i4.9754.

[26] N. Reimers and I. Gurevych, “Sentence-BERT: Sentence Embeddings using Siamese BERT-Networks,” Aug. 27, 2019, arXiv: arXiv:1908.10084. doi: 10.48550/arXiv.1908.10084.

[27] K. D. Rahadika Diana and M. L. Khodra, “IndoSBERT: Enhancing Indonesian Sentence Embeddings with Siamese Networks Fine-tuning,” in 2023 10th International Conference on Advanced Informatics: Concept, Theory and Application (ICAICTA), Lombok, Indonesia: IEEE, Oct. 2023, pp. 1–6. doi: 10.1109/ICAICTA59291.2023.10390469.

[28] O. Wibisono, A. A. Septiandri, and R. D. Najogie, “Assessing the Impact of ESG-Related News on Stock Trading in the Indonesian Market: A Text Similarity Framework Approach,” 2024. https://doi.org/10.63317/4u59w6ubzz6p .

[29] H. Thamrin, D. Oktafiani, I. I. Rasyid, and I. M. Fauzi, “Classification of SWOT Statements Employing BERT Pre-Trained Model Embedding,” J. Sist. Inf. Bisnis, vol. 14, no. 2, pp. 143–152, Apr. 2024, doi: 10.21456/vol14iss2pp143-152.

[30] A. Abuzayed and H. Al-Khalifa, “BERT for Arabic Topic Modeling: An Experimental Study on BERTopic Technique,” Procedia Comput. Sci., vol. 189, pp. 191–194, 2021, doi: 10.1016/j.procs.2021.05.096.

[31] S. Vasudeva Raju, B. Kumar Bolla, D. K. Nayak, and J. Kh, “Topic Modelling on Consumer Financial Protection Bureau Data: An Approach Using BERT Based Embeddings,” in 2022 IEEE 7th International conference for Convergence in Technology (I2CT), Mumbai, India: IEEE, Apr. 2022, pp. 1–6. doi: 10.1109/I2CT54291.2022.9824873.

[32] Nursyahrina, S. Defit, and R. Sovia, “Metode BERTopic dan LDA untuk Analisis Tren Penelitian Bidang Ilmu Komputer,” J. KomtekInfo, pp. 332–341, Sep. 2024, doi: 10.35134/komtekinfo.v11i4.580.

[33] K. M. S. Islam, R. T. Karri, S. Vegesna, J. Wu, and P. Madiraju, “Contextual Embedding-based Clustering to Identify Topics for Healthcare Service Improvement,” Apr. 18, 2025, arXiv: arXiv:2504.14068. doi: 10.48550/arXiv.2504.14068.

[34] M. Mansurova, “Topics per Class Using BERTopic,” Medium. Accessed: Jan. 22, 2024. [Online]. Available: https://towardsdatascience.com/topics-per-class-using-bertopic-252314f2640 .

[35] K. Khadijah, N. Sabilly, and F. A. Nugroho, “SENTIMENT ANALYSIS OF LEAGUE OF LEGENDS: WILD RIFT REVIEWS ON GOOGLE PLAY USING NAÏVE BAYES CLASSIFIER,” J. Ilm. Kursor, vol. 12, no. 1, pp. 23–30, Jul. 2023, doi: 10.21107/kursor.v12i01.328.

[36] I. O. Suzanti and A. Jauhari, “COMPARISON OF STEMMING AND SIMILARITY ALGORITHMS IN INDONESIAN TRANSLATED AL-QUR’AN TEXT SEARCH,” J. Ilm. Kursor, vol. 11, no. 2, p. 91, Jan. 2022, doi: 10.21107/kursor.v11i2.280.

[37] H. Axelborn and J. Berggren, “Topic Modeling for Customer Insights A Comparative Analysis of LDA and BERTopic in Categorizing Customer Calls,” UMEA University, Sweden, 2023.

[38] S. Mifrah, “Topic Modeling Coherence: A Comparative Study between LDA and NMF Models using COVID’19 Corpus,” Int. J. Adv. Trends Comput. Sci. Eng., vol. 9, no. 4, pp. 5756–5761, Aug. 2020, doi: 10.30534/ijatcse/2020/231942020.

[39] A. B. Dieng, F. J. R. Ruiz, and D. M. Blei, “Topic Modeling in Embedding Spaces,” 2019, doi: 10.48550/ARXIV.1907.04907.

[40] M. R. Hakim, “Media Pembelajaran Bahasa IMAI (Indonesia, Mandarin, Arab Dan Inggris),” skripsi, Universitas Islam Negeri Sumatera Utara, 2020. Accessed: Jan. 25, 2026. [Online]. Available: http://repository.uinsu.ac.id/13958 .

Downloads

Published

2026-07-23

Issue

Vol. 13 No. 3 (2026)

Section

Articles

Citation Check

License

Copyright (c) 2026 Nur Huda Riyantoni, Khalid Sjamsuri, Subhan Nooriansyah

Creative Commons License

This work is licensed under a Creative Commons Attribution 4.0 International License.