AZZINDANI, WILDAN AFZA (2026) PENGEMBANGAN VIRTUAL ASSISTANT DENGAN METODE RETRIEVAL-AUGMENTED GENERATION (RAG) DI BALAI BESAR SERTIFIKASI ELEKTRONIK. S1 thesis, Universitas Mercu Buana Jakarta.
|
Text (HAL COVER)
Cover.pdf Download (534kB) | Preview |
|
![[img]](https://repository.mercubuana.ac.id/style/images/fileicons/text.png) |
Text (BAB I)
Bab 1.pdf Restricted to Registered users only Download (48kB) |
|
|
Text (BAB II)
Bab 2.pdf Restricted to Registered users only Download (262kB) |
|
|
Text (BAB III)
Bab 3.pdf Restricted to Registered users only Download (110kB) |
|
|
Text (BAB IV)
Bab 4.pdf Restricted to Registered users only Download (398kB) |
|
|
Text (BAB V)
Bab 5.pdf Restricted to Registered users only Download (29kB) |
|
|
Text (DAFTAR PUSTAKA)
Daftar Pustaka.pdf Restricted to Registered users only Download (104kB) |
|
|
Text (LAMPIRAN)
Lampiran.pdf Restricted to Registered users only Download (681kB) |
Abstract
The advancement of information and communication technology has driven various government agencies to implement the Electronic-Based Government System (SPBE). One strategic unit facing challenges in digital service optimization is the Balai Besar Sertifikasi Elektronik (BSrE). With a wide user base that includes civil servants, military, and police personnel, BSrE requires a service solution capable of delivering fast and accurate responses, despite limited human resources. This study aims to develop a web-based virtual assistant system using the RetrievalAugmented Generation (RAG) method, which can automatically answer user queries based on BSrE’s internal documents as its knowledge base. The system is developed on-premise with a service architecture consisting of frontend components, messaging services, and a chatbot service. The development process follows a Research and Development (R&D) approach using the ADDIE model. System evaluation is conducted by measuring the quality of responses using evaluation metrics such as BLEU, ROUGE, and BERTScore. The results show that the developed RAG-based virtual assistant is able to generate relevant responses with good performance, supported by high user satisfaction. The system is proven to be effective in optimizing contact center services within the BSrE environment. Keywords: chatbot, contact center, RAG, LLM, artificial intelligence Perkembangan teknologi informasi dan komunikasi telah mendorong berbagai instansi pemerintah untuk mengimplementasikan Sistem Pemerintahan Berbasis Elektronik (SPBE). Salah satu unit strategis yang menghadapi tantangan dalam optimalisasi layanan secara digital adalah Balai Besar Sertifikasi Elektronik (BSrE). Dengan cakupan pengguna yang luas meliputi ASN, TNI, dan POLRI, BSrE membutuhkan solusi layanan yang mampu memberikan respons cepat dan akurat, dengan keterbatasan sumber daya manusia. Penelitian ini bertujuan untuk mengembangkan sistem virtual assistant berbasis web dengan metode RetrievalAugmented Generation (RAG), yang mampu menjawab pertanyaan pengguna secara otomatis berdasarkan dokumen internal BSrE sebagai basis pengetahuan. Sistem ini dibangun secara on-premise dengan arsitektur layanan yang terdiri atas komponen frontend, service messaging, dan service chatbot. Proses pengembangan dilakukan melalui pendekatan Research and Development (R&D) menggunakan model ADDIE. Evaluasi sistem dilakukan dengan mengukur kualitas jawaban berdasarkan metrik evaluasi seperti BLEU, ROUGE dan BERTScore. Hasil penelitian menunjukkan bahwa virtual assistant berbasis RAG yang dikembangkan mampu memberikan jawaban yang relevan. Evaluasi menggunakan metrik BLEU, ROUGE, dan BERTScore menunjukkan kinerja sistem yang baik, serta didukung oleh tingkat kepuasan pengguna yang tinggi. Sistem terbukti efektif dalam mengoptimalkan layanan contact center di lingkungan BSrE. Kata kunci: chatbot, contact center, RAG, LLM, artificial intelligence
Actions (login required)
 |
View Item |
