IMPLEMENTATION OF DISASTER MITIGATION CHATBOT USING GRAPH RETRIEVAL-AUGMENTED GENERATION (GraphRAG) FOR BPBD KABUPATEN BOGOR

Anastia Nisrina; Foni Agus Setiawan; Gibtha Fitri Laxmi

doi:10.34288/jri.v8i4.650

Authors

Anastia Nisrina Universitas Ibn Khaldun Bogor
Dr. Foni Agus Setiawan, M.Kom. BRIN Bandung
Gibtha Fitri Laxmi, S.Kom., M.Kom. Universitas Ibn Khaldun Bogor

(*) Corresponding Author

DOI:

https://doi.org/10.34288/jri.v8i4.650

Keywords:

Disaster Mitigation, Chatbot, Knowledge Graph, GraphRAG, RAGAS

Abstract

Kabupaten Bogor is among the highest disaster-risk regencies in West Java, having recorded 1,826 disaster events throughout 2025, while official mitigation guidelines remain confined to static PDF documents that are difficult to consult during emergencies. To address this gap, this study implements a disaster mitigation chatbot built on Graph Retrieval-Augmented Generation (GraphRAG), pairing Neo4j Desktop as a local graph database with Gemma3:4b (served through Ollama) as the language model, orchestrated via LangChain and exposed through a Flask web interface. The knowledge graph was built from 11 source documents provided by BPBD Kabupaten Bogor—Disaster Preparedness Pocket Books together with supplementary disaster-information documents—spanning 8 disaster types, yielding 483 nodes and 754 relationships. Each entity node carries three properties (name, description, and disaster phase) to enable more precise retrieval. System performance was assessed with the RAGAS v0.4.3 framework over a 99-question dataset over three trials, yielding average Faithfulness, Answer Relevancy, and Context Precision scores of 0.8764, 0.6773, and 0.8216, respectively. These results indicate promising contextual grounding and retrieval performance, while the lower Answer Relevancy score highlights the need for further improvement in aligning generated answers with user questions. The findings demonstrate the technical feasibility of a locally hosted GraphRAG prototype for retrieving disaster-mitigation information from BPBD Kabupaten Bogor documents. However, the evaluation does not establish operational readiness for emergency use, as comparison with conventional RAG, expert assessment, and validation with intended users remain necessary.

Downloads

Download data is not yet available.

Author Biographies

Dr. Foni Agus Setiawan, M.Kom., BRIN Bandung

Research Center for Data and Information Science

Gibtha Fitri Laxmi, S.Kom., M.Kom., Universitas Ibn Khaldun Bogor

Department of Informatics Engineering, Faculty of Engineering

References

Al-Kadi, O., & Al-Zoubi, A. (2026). From issues to insights: RAG-based explanation generation for software issue tracking systems. Proceedings of the 2026 International Conference on Software Engineering and Artificial Intelligence, 142–151.

BNPB. (2023). Buku saku tanggap tangkas tangguh menghadapi bencana. Jakarta: Badan Nasional Penanggulangan Bencana.

BNPB. (2024). Data Informasi Bencana Indonesia (DIBI). Retrieved from https://dibi.bnpb.go.id/

BPBD Kabupaten Bogor. (2023). Buku saku siaga bencana. Bogor: Badan Penanggulangan Bencana Daerah Kabupaten Bogor.

BPBD Kabupaten Bogor. (2025). Rekapitulasi data bencana tahun 2023–2025. Bogor: BPBD Kabupaten Bogor.

Chen, L., Zhang, X., & Roberts, J. (2026). A comparative evaluation of RAG architectures for cross-domain LLM applications: Design, implementation, and assessment. Journal of Enterprise Information Systems, 20(1), 89–112.

Chen, M., Tao, Z., Tang, W., Qin, T., Yang, R., & Zhu, C. (2024). Enhancing emergency decision-making with knowledge graphs and large language models. International Journal of Disaster Risk Reduction, 113, 104804.

Edge, D., Trinh, H., Cheng, N., Bradley, J., Chao, A., Mody, A., Truitt, S., & Larson, J. (2025). From local to global: A Graph RAG approach to query-focused summarization. arXiv preprint arXiv:2404.16130.

Es, S., James, J., Espinosa-Anke, L., & Schockaert, S. (2024). RAGAS: Automated evaluation of retrieval augmented generation. In Proceedings of the 18th Conference of the European Chapter of the Association for Computational Linguistics (EACL).

Gao, Y., Xiong, Y., Gao, X., Jia, K., Pan, J., Bi, Y., Dai, Y., Sun, J., & Wang, H. (2024). Retrieval-augmented generation for large language models: A survey. arXiv preprint arXiv:2312.10997.

Hikmawati, N. (2024). ASIAB chatbot for disaster mitigation learning at Madrasah Ibtidaiyah. Geosfera Indonesia, 9(2).

Neo4j, Inc. (2024). Integrating Neo4j with LangChain for GraphRAG. Neo4j Developer Blog. Retrieved from https://neo4j.com/blog/developer/

Pan, S., Luo, L., Wang, Y., Chen, C., Wang, J., & Wu, X. (2025). Unifying large language models and knowledge graphs: A roadmap. IEEE Transactions on Knowledge and Data Engineering.

Patronus AI. (2024). RAG evaluation metrics: Best practices for evaluating RAG systems. Retrieved from https://www.patronus.ai/llm-testing/rag-evaluation-metrics

Setiawan, F. A., Nisrina, A. F., & Laxmi, G. F. (2026). Knowledge Graph Dataset for Disaster Mitigation Chatbot of BPBD Kabupaten Bogor. https://hdl.handle.net/20.500.12690/RIN/YYCQ5P, RIN Dataverse, V1, UNF:6:FBw+mcI+9mqHC4DASfS1FA== [fileUNF]

Sholeh, M., Rachmawati, R. Y., & Susanti, E. (2020). Pemodelan basis data graph dengan Neo4j (studi kasus: Basis data sistem informasi penjualan pada UMKM). Jurnal Teknologi Informasi dan Terapan, 7(1), 25–30.

Sintawati, L., Rusmeilantikasari, W., Maulana, R. H., Hotimah, S., & Setiawan, E. (2025). Pemanfaatan Artificial Intelligence (AI) untuk meningkatkan mitigasi bencana banjir di Indonesia. Triwikrama: Jurnal Multidisiplin Ilmu Sosial, 9(11).

Suryanarayana, M., Rao, P. V., & Kumar, S. (2026). Real-time intelligent virtual assistant based on retrieval augmented generation (RIVA): Design and metric evaluation. International Journal of Artificial Intelligence (IJAI), 15(2), 245–258.

Urbanelli, A., Frisiello, A., Bruno, L., & Rossi, C. (2024). The ERMES chatbot: A conversational communication tool for improved emergency management and disaster risk reduction. International Journal of Disaster Risk Reduction, 112, 104792.

Vaswani, A., Shazeer, N., Parmar, N., Uszkoreit, J., Jones, L., Gomez, A. N., Kaiser, Ł., & Polosukhin, I. (2017). Attention is all you need. Advances in Neural Information Processing Systems, 30.

Wati, V., & Nugrahantoro, A. (2026). Development of a WhatsApp-based disaster mitigation chatbot using n8n workflow automation. Elkom: Jurnal Elektronika dan Komputer, 19(1), 472–482.

Wibowo, A., Amri, I., Surahmat, A., & Rusdah, R. (2025). Leveraging artificial intelligence in disaster management: A comprehensive bibliometric review. Jàmbá: Journal of Disaster Risk Studies, 17(1), a1776.

Wickramasinghe, P., Akalanka, H. M. N., Jayaprakash, S., Chathuranga, U. D. S., & Munasinghe, D. S. (2024). Chatbot-assisted disaster preparedness: A tool for educating the public on landslide-resilient building practices. In Proceedings of the 14th Annual Research Symposium, National Building Research Organisation, Sri Lanka (pp. 280–292).

Xie, C., Gao, H., Huang, Y., Xue, Z., Xu, C., & Dai, K. (2025). Leveraging the DeepSeek large model: A framework for AI-assisted disaster prevention, mitigation, and emergency response systems. Earthquake Research Advances, 5(4), 100378.

Zhang, H., et al. (2024). Knowledge graph-based RAG for disaster response: A domain-specific approach. International Journal of Disaster Risk Reduction.

Zhao, W. X., Zhou, K., Li, J., Tang, T., Wang, X., Hou, Y., Min, Y., Zhang, B., Zhang, J., Dong, Z., Du, Y., Yang, C., Chen, Y., Chen, Z., Jiang, J., Ren, R., Li, Y., Tang, X., Liu, Z., ... Wen, J.-R. (2023). A survey of large language models. arXiv preprint arXiv:2303.18223.