Rancang Bangun Sistem Informasi Konsultasi Notaris Berbasis Chatbot AI Retrieval Augmented Generation

Putra, Asraf Wisma (2026) Rancang Bangun Sistem Informasi Konsultasi Notaris Berbasis Chatbot AI Retrieval Augmented Generation. Skripsi thesis, Universitas Islam Negeri Imam Bonjol Padang.

	Text (Cover - Daftar Isi) cover-daftar isi.pdf - Published Version Download (4MB)
	Text (BAB I) BAB I.pdf - Published Version Download (301kB)
	Text (BAB III) BAB III.pdf - Published Version Download (517kB)
	Text (BAB V dan Daftar Pustaka) BAB V dan Daftar Pustaka.pdf - Published Version Download (322kB)
	Text (Full Text) Full Text.pdf - Published Version Restricted to Repository staff only Download (12MB)

Abstract

The development of Artificial Intelligence (AI) technology provides opportunities to improve information services in various fields, including notary consultation services. At the Notary Office of Pisma Putra, S.H., M.Kn., information regarding notary services, estimated costs, and administrative requirements is still primarily delivered through direct communication between clients and the notary office, resulting in limited accessibility and repeated inquiries. This research aims to design and develop a web-based Notary Consultation Information System integrated with an AI chatbot using the Retrieval-Augmented Generation (RAG) method to provide information based on the notary office's internal Knowledge Base. The research employs a Research and Development (R&D) approach with the Waterfall model for system development. The RAG implementation consists of processing internal notary documents into embeddings and retrieving relevant information using semantic search based on cosine similarity. The retrieved information is then provided as contextual input to the Gemini AI model to generate chatbot responses. System testing was conducted using Black Box Testing, RAG-based chatbot evaluation, and User Acceptance Testing (UAT). The chatbot evaluation assessed the suitability of generated responses against reference answers derived from the Knowledge Base and validated them by verifiers to determine the relevance and accuracy of the information provided. The results indicate that the developed system is capable of assisting users in obtaining notary service information based on the available knowledge sources. The implementation of RAG enables the chatbot to generate contextual and relevant responses by grounding the generation process in information retrieved from the Knowledge Base, thereby reducing the potential for unsupported or hallucinated responses.

Actions (login required)

View Item