SENTIMEN ANALISIS DATA ADUAN PELAYANAN PUBLIK DENGAN METODE NAIVE BAYES DI DUKCAPIL KABUPATEN NGAWI

ALIMIN, BINAR NUR (2026) SENTIMEN ANALISIS DATA ADUAN PELAYANAN PUBLIK DENGAN METODE NAIVE BAYES DI DUKCAPIL KABUPATEN NGAWI. S1 thesis, Universitas PGRI Madiun.

	Text HALAMAN COVER.pdf Download (380kB)
	Text ABSTRAK.pdf Download (157kB)
	Text BAB I.pdf Download (132kB)
	Text BAB II.pdf Download (422kB)
	Text BAB III.pdf Restricted to Repository staff only Download (182kB) | Request a copy
	Text BAB IV.pdf Restricted to Repository staff only Download (3MB) | Request a copy
	Text BAB V.pdf Download (102kB)
	Text DAPUS.pdf Download (95kB)
	Text LAMPIRAN.pdf Restricted to Repository staff only Download (987kB) | Request a copy

Abstract

Good public service reflects the success of local government in managing public administration, one example being the Department of Population and Civil Registration (Dukcapil) of Ngawi Regency. Public feedback through complaints serves as one measure of service quality; however, the large volume of complaints submitted in text form makes manual sentiment identification time�consuming and prone to producing less objective assessments. This study aims to design and build a website-based sentiment analysis system that classifies public service complaint data into "satisfied" and "dissatisfied" categories using the Naive Bayes method. The system was developed using the Waterfall software development method, covering the stages of requirements analysis, system design, implementation (coding), testing, and maintenance. The data used consisted of 380 public service complaint records from Dukcapil Ngawi Regency for the 2024–2026 period, with nine attributes, including complaint date, service type, complaint category, complaint description, complaint source, and satisfaction level as the class label. Testing on 76 testing data samples showed that the Naive Bayes method was able to classify complaint sentiment with an accuracy of 96.05%, precision of 95.66%, recall of 96.16%, and an F1-score of 95.89%. Functional testing using the Black Box Testing method showed that all system features functioned as required, while usability testing using the System Usability Scale (SUS) obtained an average score of 79.33, which falls into the acceptable category with a good rating. Thus, the developed system can help the Department of Population and Civil Registration of Ngawi Regency monitor and evaluate public satisfaction with public services more effectively, quickly, and objectively

Item Type:	Thesis/Skripsi/Tugas Akhir (S1)
Kata Kunci:	analisis sentimen, aduan pelayanan publik, Naive Bayes, Dukcapil, sistem berbasis website
Subjects:	T Technology > T Technology (General)
Divisions:	Fakultas Teknik > Teknik Informatika
Depositing User:	ALIMIN NUR BINAR
Date Deposited:	20 Aug 2026 07:21
Last Modified:	20 Aug 2026 07:21
URI:	http://eprint.unipma.ac.id/id/eprint/8901

Actions (login required)

View Item