Classification and Clustering of Internet Quota Sales Data Using C4.5 Algorithm and K-Means

Authors

  • Eriska Vivian Astuti ITBA-PSDKU Dian Cipta Cendikia Kotabumi
  • Asep Afandi ITBA-PSDKU Dian Cipta Cendikia Kotabumi
  • Dwi Marissa Effendi Dwimarisa89@gmail.com

DOI:

https://doi.org/10.26555/jiteki.v9i2.25970

Keywords:

C4.5 Algorithm, K-Means, Data mining, Quota

Abstract

The number of restrictions or limits on internet use is known as internet quota. When you use internet data for a short time, you usually run out of bandwidth. In the Candimas South Abung area, many quotas have been sold in various variants. Visitors to quota outlets have access to various kinds of quota references that they can buy. Apart from guaranteeing the quality of the quotas sold, sales always increase every year, especially in the various quota variants. Based on quota data for 2019 to 2022. This study aims to analyze internet quota sales statistics in the Candimas area between 2019 and 2022. In 2021-2022 the classification produces an accuracy of up to 100% where the best-selling data dominates while clustering remains at the same figure, namely 19 data are very salable, 43 data are lacking sold, and 178 data did not sell. We use the C4.5 classification algorithm and K-Means clustering to identify patterns in the data and provide insight into which brand quotas are the most popular. Our findings can help Xena Cell counter owners make informed decisions about which quota to add or remove to optimize sales and minimize losses.

Downloads

Published

2023-04-20

Issue

Section

Articles