Hadith Topic Modeling Based on Theme Similarity with the BERTopic Algorithm
Article Sidebar
Main Article Content
Abstract
This study aims to develop a hadith topic modeling system based on thematic similarity using the BERTopic algorithm to support more structured and efficient Islamic studies. The research method uses a data exploration-based approach with the BERTopic algorithm that combines BERT semantic representation, UMAP dimensionality reduction, and HDBSCAN clustering. The dataset used is sourced from Kaggle containing thousands of hadith texts in English from various books, which are processed through the stages of data pre-processing, embedding transformation, and evaluation using topic coherence metrics and interactive visualization. The results show that BERTopic successfully identified 292 topics with clear visualization through the Intertopic Distance Map and Topic Similarity Heatmap; however, approximately 55.4% of documents are categorized as outliers due to high semantic variation and limited context. Despite the challenges in handling outlier documents, this study proves that BERTopic is effective in clustering hadith based on thematic similarity and producing semantically coherent topics. The resulting model can be the basis for the development of a hadith recommendation system, classification of da'wah content, and a theme-based religious learning curriculum, as well as opening up opportunities for the application of NLP technology in Islamic studies more broadly.
Downloads
Article Details
Section
This work is licensed under a Creative Commons Attribution 4.0 International License.
You are free to:
Share — copy and redistribute the material in any medium or format
Adapt — remix, transform, and build upon the material for any purpose, even commercially.
The licensor cannot revoke these freedoms as long as you follow the license terms.
Under the following terms:
Attribution — You must give appropriate credit, provide a link to the license, and indicate if changes were made. You may do so in any reasonable manner, but not in any way that suggests the licensor endorses you or your use.
No additional restrictions — You may not apply legal terms or technological measures that legally restrict others from doing anything the license permits.
Notices:
You do not have to comply with the license for elements of the material in the public domain or where your use is permitted by an applicable exception or limitation.
No warranties are given. The license may not give you all of the permissions necessary for your intended use. For example, other rights such as publicity, privacy, or moral rights may limit how you use the material.
How to Cite
References
[1] R. Aban, “Sunnah (Hadith) Dataset.” 2022.
[2] DM Blei, AY Ng, and MI Jordan, “Latent Dirichlet Allocation,” J. Mach. Learn. Res. , vol. 3, pp. 993–1022, 2003.
[3] J. Devlin, M.-W. Chang, K. Lee, and K. Toutanova, “BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding,” ArXiv Prepr. , 2018.
[4] M. Grootendorst, “BERTopic: Neural topic modeling with a class-based TF-IDF procedure,” ArXiv Prepr. , 2022.
[5] S. Al-Shargi, “LDA Topic Modeling of the Holy Quran Text.” Nov. 2024. [Online]. Available: https://medium.com/@alshargi.usa/lda-topic-modeling-of-the-holy-quran-text-b43d3d7348c9
[6] D. Kumar, “Topic Modeling with BERTopic: A Comprehensive Analysis of Modern NLP Applications,” Int. J. Mach. Learn. Comput. , vol. 15, no. 1, pp. 12–20, Jan. 2025.
[7] A. Hassan, M. Ibrahim, and S. Khalil, “Hadith Classification using Machine Learning Techniques According to its Reliability,” Int. J. Adv. Comput. Sci. Appl. , vol. 10, no. 12, pp. 256–264, 2019.
[8] K. Al-Rashid, H. Mohammad, and F. Ahmed, “Text categorization in Quran and Hadith: Overcoming the interrelation challenges using machine learning and term weighting,” J. King Saud Univ. - Comput. Inf. Sci. , vol. 32, no. 8, pp. 914–923, Oct. 2020, doi: 10.1016/j.jksuci.2019.04.003.
[9] IK Alshammari, E. Atwell, and MA Alsalka, “Topic Modeling for Hadith Corpus: A Comparison of Latent Dirichlet Allocation (LDA), Non-Negative Matrix Factorization (NMF), and BERTopic with AraBERT, XLM-R, MARBERT, and CAMeLBERT,” Int. J. Islam. Appl. Comput. Sci. Technol. , vol. 11, no. 4, pp. 9–16, Dec. 2023.
[10] N. Alqahtani, F. Alotaibi, and M. Alsaeed, “Employing a Multilingual Transformer Model for Segmenting Unpunctuated Arabic Text,” Appl. Sci. , vol. 12, no. 20, p. 10559, Oct. 2022, doi: 10.3390/app122010559.
[11] L. Abdullah, K. Yaseen, and H. Naji, “AraBERT transformer model for Arabic comments and reviews analysis,” Neural Comput. Appl. , vol. 34, no. 8, pp. 6271–6285, Mar. 2022, doi: 10.1007/s00521-021-06782-4.
[12] S. Aftar, L. Gagliardelli, AE Ganadi, F. Ruozzi, and S. Bergamaschi, “RoBERT2VecTM: A Novel Approach for Topic Extraction in Islamic Studies,” in Findings of the Association for Computational Linguistics: EMNLP 2024 , 2024. doi: 10.18653/v1/2024.findings-emnlp.534.
[13] 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.
[14] T. Zerrouki, A. Balla, and A. Lakhouaja, “Arabic natural language processing for Qur'anic research: a systematic review,” vol. 56, no. 2, pp. 1175–1207, Feb. 2023, doi: 10.1007/s10462-022-10313-2.
[15] K. Jamshaid, H. Farooq, and MT Siddique, “Topic Modeling of Quranic Verses using Latent Dirichlet Allocation with English Language: Topic Modeling using LDA,” VFAST Trans. Softw. Eng. , vol. 12, no. 4, pp. 239–251, Dec. 2024, doi: 10.21015/vtse.v12i4.1946.
[16] AMA Egov and T. Grok, “The Development of Topic Modeling for Hadith Studies,” J. Islam. Stud. , vol. 2021, 2021.
[17] MA Mosa, "An Exhaustive Literature Review of Hadith Text Mining," ACM Trans. Asian Low-Resort. Lang. Inf. Process. , vol. 22, no. 7, pp. 1–25, Jul. 2023, doi: 10.1145/3588315.
[18] Herwinsyah and others, “Topic Modeling in the Qur'an Using the BERTopic Library on the BERT Language Model,” vol. 14, no. 2, 2023.
[19] F. Alotaibi, M. Alharbi, and S. Alnasser, “ArabianGPT: Native Arabic GPT-based Large Language Model,” ArXiv Prepr. ArXiv240215313, Feb. 2024.
[20] R. Baly, G. Hajj, and W. El-Hajj, “Benchmarking Transformer-based Language Models for Arabic Sentiment and Sarcasm Detection,” in Proceedings of the 6th Arabic Natural Language Processing Workshop , Association for Computational Linguistics, Apr. 2021, pp. 104–114.