Penerapan Algoritma C45 Untuk Mengklasifikasi Kepatuhan Pelanggan Indihome Dalam Membayar Tagihan

Fadhila, Indrayani (2021) Penerapan Algoritma C45 Untuk Mengklasifikasi Kepatuhan Pelanggan Indihome Dalam Membayar Tagihan. Technical Report. Pustakawan, Perpustakaan Institut Teknologi Telkom Purwokerto. (Unpublished)

	Text COVER .pdf Download (100kB)
	Text ABSTRACT.pdf Download (80kB)
	Text ABSTRAK.pdf Download (83kB)
	Text BAB I.pdf Download (293kB)
	Text BAB II.pdf Download (132kB)
	Text BAB III.pdf Restricted to Registered users only Download (274kB) | Request a copy
	Text BAB IV.pdf Download (83kB)
	Text DAFTAR PUSTAKA.pdf Download (95kB)
	Text LAMPIRAN.pdf Restricted to Registered users only Download (608kB) | Request a copy

Abstract

The internet has become a necessity in supporting access to information and communication. In Indonesia, there is one Internet service provider company, namely PT Telkom Indonesia. PT Telkom Indonesia has one product, namely IndiHome (Indonesia Digital Home) in the form of Wi-Fi-based internet services. IndiHome service users have regulations in bill payments, namely payments are made on the twentieth of each month. IndiHome employees can view customer bill payment information through the e-info payment system, but the system does not contain information on customer compliance in complying with payment regulations. So it is necessary to classify customer compliance in paying bills. The purpose of this study is to look at the classification of customer compliance in paying bills so as to minimize the amount of punishment that will be given to customers. Data collection was carried out in August – September at Telkom Witel Purwokerto. The C4.5 algorithm was chosen as a method for classifying data because it can produce an efficient decision tree and acceptable accuracy results. The tools used for classification use RapidMiner software which is an application for analyzing data mining. The data used is 64 IndiHome customer data from Banyumas district. There are 4 attributes used, namely "payment achievement", "payment date", "due date" and "description". Based on the results of the decision tree, the attribute that most influences customer compliance is payment achievement which gets the highest Gain value of 1.217022083, and produces 2 rules. Suggestions for the Telkom office to add a table of customer compliance information in paying bills in the e-Info payment system. Keywords: C4.5 Algorithm, Customer Compliance, RapidMiner.

Actions (login required)

View Item