| dc.contributor.advisor | Aziezah, Nur | |
| dc.contributor.author | DAMARA, RAFLI | |
| dc.date.accessioned | 2026-07-27T04:42:42Z | |
| dc.date.available | 2026-07-27T04:42:42Z | |
| dc.date.issued | 2026 | |
| dc.identifier.uri | http://repository.ipb.ac.id/handle/123456789/175923 | |
| dc.description.abstract | Layanan informasi publik di BRMP Penerapan saat ini memiliki
keterbatasan jam operasional dan penanganan manual. Penelitian ini bertujuan
mengembangkan aplikasi chatbot sebagai solusi otomatisasi layanan yang responsif
dan beroperasi 24 jam. Sistem dibangun menggunakan pendekatan Natural
Language Processing dengan arsitektur hibrida, yaitu mengombinasikan algoritma
Deep Learning untuk klasifikasi intent pengguna dan rule-based untuk
pengendalian respons akhir. Pengembangan menggunakan Python, framework
Flask, serta metode Waterfall. Evaluasi sistem meliputi tiga tahap pengujian.
Pertama, pengujian Black Box mencapai tingkat keberhasilan 100%. Kedua,
evaluasi performa model Deep Learning dalam mengklasifikasikan intent
membuktikan keandalan sistem dengan tingkat akurasi sebesar 94,59%. Ketiga,
Business Acceptance Testing mengonfirmasi bahwa aplikasi chatbot ini telah
diterima dan sukses memenuhi kebutuhan operasional bisnis instansi. | |
| dc.description.abstract | Public information services at BRMP Penerapan currently encounter
constraints regarding operational hours and manual handling. This research aims to
develop a chatbot application as a responsive automation solution operating 24
hours. The system is built using a Natural Language Processing approach with a
hybrid architecture, combining a Deep Learning algorithm for user intent
classification and a rule-based method for final response control. The development
utilizes Python, the Flask framework, and the Waterfall method. The system
evaluation comprises three testing phases. First, Black Box testing achieves a 100%
success rate. Second, performance evaluation of the Deep Learning model in
classifying intents demonstrates system reliability with an accuracy rate of 94.59%.
Third, Business Acceptance Testing confirms that the chatbot application is well
accepted and successfully meets the agency's operational business needs. | |
| dc.description.sponsorship | | |
| dc.language.iso | id | |
| dc.publisher | IPB University | id |
| dc.title | Aplikasi Chatbot Rule-based Menggunakan NLP dan Framework Flask Untuk Layanan Informasi Publik BRMP Penerapan | id |
| dc.title.alternative | Rule-Based Chatbot Application Using NLP and Flask Framework for BRMP's "Penerapan" Public Information Service | |
| dc.type | Tugas Akhir | |
| dc.subject.keyword | business acceptance testing | id |
| dc.subject.keyword | Chatbot | id |
| dc.subject.keyword | deep learning | id |
| dc.subject.keyword | flask | id |
| dc.subject.keyword | natural language processing | id |
| dc.subtype | Undergraduate Theses | |