Pengembangan Aplikasi Informasi Obat Berbasis Retrieval-Augmented Generation dengan Grounded Knowledge dari Buku Daftar Obat Indonesia (DOI)

Wijaya, Gabriella Erlinda (2026) Pengembangan Aplikasi Informasi Obat Berbasis Retrieval-Augmented Generation dengan Grounded Knowledge dari Buku Daftar Obat Indonesia (DOI). Other thesis, Institut Teknologi Sepuluh Nopember.

Text 5027221018-Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only Download (11MB) | Request a copy

Abstract

Akses terhadap informasi obat yang cepat, akurat, dan kontekstual merupakan kebutuhan penting bagi tenaga kesehatan maupun masyarakat umum. Salah satu sumber rujukan resmi di Indonesia adalah Daftar Obat Indonesia (DOI) yang masih banyak tersedia dalam format buku fisik, sehingga menyulitkan pencarian, kurang portabel, dan rentan rusak. Penelitian ini bertujuan untuk mengembangkan aplikasi mobile berbasis Android yang mampu menyediakan layanan informasi obat dengan memanfaatkan pendekatan Retrieval-Augmented Generation (RAG). Data obat diperoleh melalui digitalisasi buku DOI yang selanjutnya digunakan sebagai grounded knowledge untuk sistem. Aplikasi yang dibangun dilengkapi dengan fitur pencarian cerdas (smart search) yang toleran terhadap variasi input, serta tanya jawab berbasis AI yang didukung oleh Large Language Model (LLM). Metodologi penelitian meliputi studi literatur, analisis kebutuhan, perancangan arsitektur sistem, implementasi, serta pengujian yang mencakup akurasi pencarian, ketepatan hasil RAG, dan usability testing. Hasil penelitian menunjukkan bahwa fitur smart search berbasis fuzzy matching mencapai skor Mean Reciprocal Rank (MRR) sebesar 0,96 pada pencarian obat sediaan. Selain itu, mekanisme RAG berhasil meningkatkan nilai groundedness jawaban hingga 38,12% sekaligus menekan tingkat halusinasi secara signifikan. Dari sisi pengalaman pengguna, aplikasi memperoleh skor System Usability Scale (SUS) sebesar 89,25% yang termasuk dalam kategori Grade A+. Hal ini membuktikan bahwa aplikasi yang dikembangkan efektif, akurat, dan mudah digunakan, sehingga dapat menjadi solusi digitalisasi data farmasi yang mendukung akses informasi obat di Indonesia.
=================================================================================================================================
Access to fast, accurate, and contextual medication information is a critical necessity for both healthcare professionals and the general public. One of the official references in Indonesia is the Daftar Obat Indonesia (DOI), which is still predominantly available in physical book format, leading to difficulties in searching, lack of portability, and vulnerability to damage. This research aims to develop an Android-based mobile application that provides medication information services by utilizing the Retrieval-Augmented Generation (RAG) approach. Medication data were obtained through the digitalization of the DOI book, which subsequently served as grounded knowledge for the system. The developed application features a smart search function tolerant of input variations and an AI-based question-and-answering system supported by a Large Language Model (LLM) to generate more contextual responses. The research methodology encompasses literature review, requirements analysis, system architecture design, application implementation, and testing, which includes search accuracy, RAG precision, and usability testing. The results indicate that the fuzzy matching-based smart search feature achieved a Mean Reciprocal Rank (MRR) score of 0.96 for dosage form searches. Furthermore, the RAG mechanism successfully increased the groundedness of answers by 38,12% while simultaneously reducing hallucination rates. From a user experience perspective, the application obtained a System Usability Scale (SUS) score of 89.25%, categorizing it as Grade A+. This proves that the system is effective, accurate, and user-friendly, providing a viable solution for pharmaceutical data digitalization and supporting broader access to medication information in Indonesia.

Item Type:	Thesis (Other)
Uncontrolled Keywords:	Retrieval-Augmented Generation (RAG), Grounded Knowledge, Daftar Obat Indonesia (DOI), Smart Search, LLM, Aplikasi Mobile, Retrieval-Augmented Generation, Grounded Knowledge, Indonesian Drug Directory, Smart Search, Large Language Model, Mobile Application
Subjects:	T Technology > T Technology (General) T Technology > T Technology (General) > T57.5 Data Processing T Technology > T Technology (General) > T58.5 Information technology. IT--Auditing T Technology > T Technology (General) > T58.62 Decision support systems
Divisions:	Faculty of Intelligent Electrical and Informatics Technology (ELECTICS) > Information Technology > 59201-(S1) Undergraduate Thesis
Depositing User:	Gabriella Erlinda Wijaya
Date Deposited:	21 Jan 2026 05:47
Last Modified:	21 Jan 2026 05:47
URI:	http://repository.its.ac.id/id/eprint/129934

Actions (login required)

View Item