Klasifikasi Customer Churn pada Telekomunikasi Industri Untuk Retensi Pelanggan Menggunakan Algoritma C4.5

Stevan Desena Damanik (Universitas Sriwijaya, Palembang)
Muhammad Ihsan Jambak (Universitas Sriwijaya, Palembang)

KLIK: Kajian Ilmiah Informatika dan Komputer
Vol. 3 No. 6 (2023): Juni 2023

Ignorance of telecommunications companies regarding the reasons and characteristics of customer churn causes telecommunications companies to suffer huge losses. This makes customer churn a big problem for telecommunications companies. This study uses data mining with classification techniques as a solution to analyze customer churn characteristics. This research will use Rapid Miner and the C4.5 algorithm to carry out the data mining process. . The purpose of this research is to find out what are the characteristics of customer churn so that companies can make policies that can retain customers and increase customer retention. This research is based on CRISP-DM. Data taken from kaggle.com with 21 attributes and 7034 rows of data and data preparation will be carried out. From the research results it is known that there are 5 attributes that have a considerable influence on customer churn, namely contracts, InternetService, TotalChares, tenure, PaperlessBilling, MultipleLines, StreamingMovies. And from the results of this study has an accuracy rate of 79.53%.