Optimization Optimization of Backward Elimination for Classification of Customer Satisfaction Using the k-nearest neighbor (k-NN) and Naive Bayes Algorithm

Technomedia Journal

View Publication Info
 
 
Field Value
 
Title Optimization Optimization of Backward Elimination for Classification of Customer Satisfaction Using the k-nearest neighbor (k-NN) and Naive Bayes Algorithm
Optimasi Backward Elimination untuk Klasifikasi Kepuasan Pelanggan Menggunakan Algoritme k-nearest neighbor (k-NN) and Naive Bayes
 
Creator Yunitasari
Hopipah, Hopi Siti
Mayasari, Rini
 
Description Maintaining customer satisfaction is a big challenge for companies. One effort that can be done is to provide the best service to customers based on the most influential aspects. In this study, the optimization of the Backward Elimination feature in the classification of customer satisfaction using the k-NN and Naïve Bayes algorithm. The use of the Backward Elimination feature aims to increase accuracy and reduce the number of less influential attributes. As a result, it can be seen that the best modeling without Backward Elimination is the Naïve Bayes algorithm with an accuracy of 99.04% and an AUC value of 1. While the application of Backward Elimination works more optimally on the k-NN algorithm with an increase of 33.74% to 97.28% with AUC 0.996. This shows that the performance of the Backward Elimination feature is effective in optimizing the classification of customer satisfaction and can reduce the less influential attributes.
Mempertahankan kepuasan pelanggan merupakan sebuah tantangan besar bagi perusahaan. Salah satu upaya yang dapat dilakukan adalah memberikan pelayanan terbaik terhadap pelanggan berdasarkan aspek yang paling berpengaruh. Pada penelitian ini dilakukan optimasi fitur Backward Elimination pada klasifikasi kepuasan pelanggan dengan algoritme k-NN dan Naïve Bayes. Penggunaan fitur Backward Elimination bertujuan meningkatkan akurasi dan mengurangi jumlah atribut yang kurang berpengaruh. Hasilnya, dapat diketahui bahwa pemodelan terbaik tanpa Backward Elimination adalah algoritme Naïve Bayes dengan akurasi 99.04% dan nilai AUC mencapai 1. Sedangkan penerapan Backward Elimination bekerja lebih optimal pada algoritme k-NN dengan peningkatan sebesar 33.74% menjadi 97.28% dengan AUC 0.996. Hal ini menunjukkan bahwa kinerja fitur Backward Elimination efektif dalam optimasi klasifikasi kepuasan pelanggan dan dapat mengurangi atribut yang kurang berpengaruh.
 
Publisher iLearning Journal Center (iJC)
 
Date 2021-07-16
 
Type info:eu-repo/semantics/article
info:eu-repo/semantics/publishedVersion
 
Format application/pdf
 
Identifier http://ijc.ilearning.co/index.php/TMJ/article/view/1531
10.33050/tmj.v6i1.1531
 
Source Technomedia Journal; Vol. 6 No. 1 (2021): TMJ (Technomedia Journal); 99-110
Technomedia Journal; Vol 6 No 1 (2021): TMJ (Technomedia Journal); 99-110
2528-6544
2620-3383
10.33050/tmj.v6i1
 
Language ind
 
Relation http://ijc.ilearning.co/index.php/TMJ/article/view/1531/479
 
Rights Hak Cipta (c) 2021 Yunitasari, Hopi Siti Hopipah, Rini Mayasari
https://creativecommons.org/licenses/by/4.0/
 

Contact Us

The PKP Index is an initiative of the Public Knowledge Project.

For PKP Publishing Services please use the PKP|PS contact form.

For support with PKP software we encourage users to consult our wiki for documentation and search our support forums.

For any other correspondence feel free to contact us using the PKP contact form.

Find Us

Twitter

Copyright © 2015-2018 Simon Fraser University Library