Laporan Kerja Praktik 23 Februari - 19 Juni 2026 di Otoritas Jasa Keuangan

Putri, Dinda Ayu Ningratu and Anam, Choirul and Riavinola, Annisa Salma (2026) Laporan Kerja Praktik 23 Februari - 19 Juni 2026 di Otoritas Jasa Keuangan. Project Report. [s.n.], [s.l.]. (Unpublished)

Text 5025231059_5025231145_5025231312-Project_Report.pdf - Accepted Version Restricted to Repository staff only Download (849kB) | Request a copy

Abstract

Otoritas Jasa Keuangan (OJK) merupakan lembaga independen yang bertugas mengatur dan mengawasi sektor jasa keuangan di Indonesia. Meskipun website OJK telah menyediakan konten edukasi yang cukup lengkap, masyarakat masih kesulitan mengakses informasi karena konten tersebar di banyak halaman dan disajikan dalam format statis yang kurang interaktif. Selain itu, tim internal OJK menghadapi beban kerja manual yang tinggi akibat belum adanya sistem otomasi dalam menangani pertanyaan publik yang berulang. Proyek ini bertujuan membangun asisten virtual berbasis percakapan bernama ROJAK (Robot Penjawab Kontak OJK) yang diintegrasikan ke dalam website resmi OJK di bawah koordinasi tim Edukasi dan Perlindungan Konsumen (EPK). Sistem ini dikembangkan menggunakan arsitektur Agentic Retrieval Augmented Generation (RAG) dengan kerangka kerja Next.js, xvi basis data relasional NeonDB, serta database vektor Pinecone sebagai fondasi teknisnya. Hasil pengujian fungsional dan performa menunjukkan bahwa ROJAK mampu berfungsi sebagai pintu gerbang percakapan tunggal yang responsif dengan rata-rata waktu respons yang cepat dan ketahanan tinggi terhadap ancaman siber. Sistem ini juga dilengkapi Admin Dashboard komprehensif untuk memantau metrik performa dan tren pertanyaan publik secara real-time. Secara keseluruhan, ROJAK telah memenuhi kriteria Minimum Viable Product (MVP) yang layak untuk diimplementasikan pada skala lebih luas. Meski demikian, masih terdapat kesenjangan pada indikator Intent Coverage akibat keterbatasan cakupan dokumen, sehingga pengembangan ke depan perlu difokuskan pada perluasan basis pengetahuan dan optimasi retrieval pipeline.
=======================================================================================================================================
The Indonesian Financial Services Authority (OJK) is an independent institution responsible for regulating and supervising the financial services sector in Indonesia. Although the OJK website provides comprehensive educational content, the public still encounters difficulties in accessing information because the content is distributed across numerous webpages and presented in a static, less interactive format. Furthermore, the internal OJK team faces a significant manual workload due to the absence of an automated system for handling repetitive public inquiries.This project aims to develop a conversational virtual assistant named **ROJAK (Robot Penjawab Kontak OJK)**, which is integrated into the official OJK website under the coordination of the Education and Consumer Protection (EPK) Division. The system is developed using an Agentic Retrieval-Augmented Generation (RAG) architecture, with Next.js as the application framework, NeonDB as the relational database, and Pinecone as the vector database forming its technical foundation.The results of functional and performance testing demonstrate that ROJAK effectively serves as a responsive single conversational gateway, delivering fast average response times while maintaining strong resilience against cyber threats. The system is also equipped with a comprehensive Admin Dashboard that enables real-time monitoring of performance metrics and public inquiry trends.Overall, ROJAK has successfully fulfilled the criteria of a Minimum Viable Product (MVP), making it suitable for broader implementation. Nevertheless, a gap remains in the Intent Coverage metric due to the limited scope of the underlying document corpus. Therefore, future development should focus on expanding the knowledge base and optimizing the retrieval pipeline.

Item Type:	Monograph (Project Report)
Uncontrolled Keywords:	Chatbot, Agentic Retrieval-Augmented Generation (Agentic RAG), Financial Literacy, Financial Services Authority (OJK), Dashboard Admin, Chatbot, Agentic RAG, Literasi Keuangan, Otoritas Jasa Keuangan (OJK), Admin Dashboard
Subjects:	T Technology > T Technology (General) > T58.62 Decision support systems T Technology > T Technology (General) > T59.7 Human-machine systems.
Divisions:	Faculty of Intelligent Electrical and Informatics Technology (ELECTICS) > Informatics Engineering > 55201-(S1) Undergraduate Thesis
Depositing User:	Choirul Anam
Date Deposited:	29 Jun 2026 02:16
Last Modified:	29 Jun 2026 02:16
URI:	http://repository.its.ac.id/id/eprint/134119

Actions (login required)

View Item