IMPLEMENTASI METODE NAIVE BAYES DAN KNN UNTUK PREDIKSI TINGKAT KEPUASAN PELANGGAN CS FLOAT

NUR HINDUN SYA’ADA SIREGAR, NPM 2209100094 (2026) IMPLEMENTASI METODE NAIVE BAYES DAN KNN UNTUK PREDIKSI TINGKAT KEPUASAN PELANGGAN CS FLOAT. Skripsi thesis, Universitas Labuhanbatu.

	Text COVER.pdf Download (726kB)
	Text BAB I.pdf Download (180kB)
	Text BAB II.pdf Download (348kB)
	Text BAB III.pdf Download (516kB)
	Text BAB IV.pdf Download (746kB)
	Text BAB V.pdf Download (98kB)
	Text DAFTAR PUSTAKA.pdf Download (160kB)

Abstract

Penelitian ini diawali dengan latar belakang pentingnya memahami tingkat kepuasan pelanggan terhadap produk minuman CS Float sebagai upaya meningkatkan daya saing di tengah ketatnya persaingan bisnis minuman, serta perlunya pendekatan berbasis data untuk menghasilkan analisis yang objektif. Dalam landasan teori, dijelaskan bahwa data mining dan machine learning, khususnya metode klasifikasi seperti Naive Bayes dan K-Nearest Neighbor (KNN), mampu mengolah data pelanggan menjadi informasi yang berguna untuk prediksi kepuasan. Pada tahap analisis dan perancangan, dilakukan proses pengolahan data mulai dari pembersihan data, transformasi ke bentuk kategorikal dan numerik, hingga pembagian data menjadi data training dan testing untuk membangun model klasifikasi. Selanjutnya, model dirancang menggunakan aplikasi Orange dengan memanfaatkan berbagai widget seperti file, data preprocessing, classification, dan evaluation untuk menghasilkan alur analisis yang sistematis. Hasil dan pembahasan menunjukkan bahwa kedua metode yang digunakan mampu menghasilkan performa yang sangat baik dengan tingkat akurasi, presisi, dan recall mencapai 100%, serta mampu mengklasifikasikan data kepuasan pelanggan secara tepat. Selain itu, hasil klasifikasi pada data testing menunjukkan dominasi kategori puas dibandingkan tidak puas, yang mengindikasikan bahwa sebagian besar pelanggan memiliki persepsi positif terhadap produk. Berdasarkan keseluruhan proses dan hasil yang diperoleh, dapat disimpulkan bahwa metode Naive Bayes dan KNN sama-sama efektif dalam memprediksi tingkat kepuasan pelanggan, sehingga keduanya dapat dijadikan sebagai alternatif model dalam pengambilan keputusan berbasis data di bidang bisnis. Kata Kunci : Data Mining, K-Nearest Neighbor, Kepuasan Pelanggan, Naive Bayes, Machine Learning. ================================================================================================ This research begins with the background of the importance of understanding the level of customer satisfaction with CS Float beverage products as an effort to increase competitiveness amidst the tight competition in the beverage business, as well as the need for a data-driven approach to produce objective analysis. In the theoretical basis, it is explained that data mining and machine learning, especially classification methods such as Naive Bayes and K-Nearest Neighbor (KNN), are able to process customer data into useful information for satisfaction prediction. In the analysis and design stage, the data processing process is carried out starting from data cleaning, transformation into categorical and numerical forms, to dividing the data into training and testing data to build a classification model. Next, the model is designed using the Orange application by utilizing various widgets such as files, data preprocessing, classification, and evaluation to produce a systematic analysis flow. The results and discussion show that both methods used are able to produce excellent performance with accuracy, precision, and recall levels reaching 100%, and are able to classify customer satisfaction data accurately. In addition, the classification results on the testing data show a dominance of the satisfied category compared to dissatisfied, which indicates that most customers have a positive perception of the product. Based on the overall process and results obtained, it can be concluded that the Naive Bayes and KNN methods are equally effective in predicting customer satisfaction levels, so both can be used as alternative models in data-based decision making in the business sector. Keywords : Data Mining, K-Nearest Neighbor, Customer Satisfaction, Naive Bayes, Machine Learning.

Item Type:	Thesis (Skripsi)
Uncontrolled Keywords:	Data Mining, K-Nearest Neighbor, Kepuasan Pelanggan, Naive Bayes, Machine Learning. =========================================== Data Mining, K-Nearest Neighbor, Customer Satisfaction, Naive Bayes, Machine Learning.
Subjects:	H Social Sciences > HF Commerce Q Science > QA Mathematics > QA76 Computer software T Technology > T Technology (General)
Divisions:	Fakultas Sains Dan Teknologi > Sistem Informasi
Depositing User:	Unnamed user with email repository@ulb.ac.id
Date Deposited:	27 Apr 2026 03:26
Last Modified:	27 Apr 2026 03:30
URI:	http://repository.ulb.ac.id/id/eprint/2165

Actions (login required)

View Item