ANALISIS SENTIMEN KEPUASAN CUSTOMER TERHADAP LAYANAN SERVIS HANDPHONE MENGGUNAKAN METODE NAIVE BAYES (STUDI KASUS JW PONSEL RANTAUPRAPAT)

RANI RAHMA WATI, NPM 2109100066 (2025) ANALISIS SENTIMEN KEPUASAN CUSTOMER TERHADAP LAYANAN SERVIS HANDPHONE MENGGUNAKAN METODE NAIVE BAYES (STUDI KASUS JW PONSEL RANTAUPRAPAT). Skripsi thesis, Universitas Labuhanbatu.

	Text COVER.pdf Download (1MB)
	Text BAB I.pdf Download (240kB)
	Text BAB II.pdf Download (405kB)
	Text BAB III.pdf Download (405kB)
	Text BAB IV.pdf Download (2MB)
	Text BAB V.pdf Download (189kB)
	Text DAFTAR PUSTAKA.pdf Download (161kB)

Abstract

Penelitian ini dilakukan untuk menganalisis sentimen kepuasan pelanggan terhadap layanan servis handphone di JW Ponsel Rantauprapat menggunakan Algoritma Naïve Bayes, yang dikenal efektif untuk klasifikasi berbasis probabilitas pada data kategorikal. Landasan teori mengacu pada prinsip kerja Naïve Bayes yang menghitung probabilitas kemunculan atribut secara independen, sehingga mampu mengelompokkan opini menjadi sentimen positif atau negatif berdasarkan kombinasi variabel layanan seperti kecepatan servis, harga servis, dan kualitas servis. Analisis dan perancangan sistem dimulai dengan pengumpulan dan Preprocessing Data ulasan pelanggan, pembagian menjadi data latih dan data uji, lalu pemodelan di RapidMiner menggunakan Operator klasifikasi dan evaluasi. Hasil klasifikasi terhadap 80 data pelanggan menunjukkan 57 data diprediksi positif dan 23 negatif, dengan kombinasi atribut “Cepat – Murah – Bagus” cenderung positif, sedangkan “Lambat – Mahal – Kurang Bagus” cenderung negatif. Evaluasi performa menghasilkan akurasi 97,50% dengan nilai presisi dan recall kelas positif 98,25% dan kelas negatif 95,65%, serta hanya terdapat masing-masing satu kesalahan pada False Positive dan False Negative. Kesimpulannya, model Naïve Bayes yang dibangun terbukti akurat, andal, dan konsisten, sehingga dapat menjadi alat bantu strategis bagi manajemen JW Ponsel dalam memantau kepuasan pelanggan dan merumuskan perbaikan layanan secara berbasis data. Kata Kunci: Analisis Sentimen, Naïve Bayes, Klasifikasi, Kepuasan Pelanggan, RapidMiner ================================================================================================= This study was conducted to analyze customer satisfaction sentiment towards mobile phone service at JW Ponsel Rantauprapat using the Naïve Bayes algorithm, which is known to be effective for probability-based classification on categorical data. The theoretical basis refers to the working principle of Naïve Bayes which calculates the probability of attribute occurrence independently, so that it is able to group opinions into positive or negative sentiments based on a combination of service variables such as service speed, service price, and service quality. The analysis and design of the system began with the collection and preprocessing of customer review data, division into training and test data, then modeling in RapidMiner using classification and evaluation Operators. The classification results of 80 customer data showed 57 data predicted as positive and 23 as negative, with the attribute combination "Fast - Cheap - Good" tending to be positive, while "Slow - Expensive - Less Good" tended to be negative. The Performance evaluation produced an accuracy of 97.50% with a precision and recall value of 98.25% for the positive class and 95.65% for the negative class, and there was only one error each in False Positive and False Negative. In conclusion, the Naïve Bayes model built is proven to be accurate, reliable, and consistent, so it can be a strategic tool for JW Ponsel management in monitoring customer satisfaction and formulating data-based service improvements. Keywords: Sentiment Analysis, Naïve Bayes, Classification, Customer Satisfaction, RapidMiner

Item Type:	Thesis (Skripsi)
Uncontrolled Keywords:	Analisis Sentimen, Naïve Bayes, Klasifikasi, Kepuasan Pelanggan, RapidMiner=============Sentiment Analysis, Naïve Bayes, Classification, Customer Satisfaction, RapidMiner
Subjects:	Q Science > QA Mathematics > QA75 Electronic computers. Computer science T Technology > T Technology (General) Z Bibliography. Library Science. Information Resources > Z665 Library Science. Information Science Z Bibliography. Library Science. Information Resources > ZA Information resources > ZA4050 Electronic information resources
Divisions:	Fakultas Sains Dan Teknologi > Sistem Informasi
Depositing User:	Unnamed user with email repository@ulb.ac.id
Date Deposited:	31 Oct 2025 03:12
Last Modified:	31 Oct 2025 03:12
URI:	http://repository.ulb.ac.id/id/eprint/1890

Actions (login required)

View Item