Customer Churn Prediction Using Machine Learning Methods: A Comparative Analysis Makine Öğrenmesi Yöntemleri Kullanılarak Müşteri Kaybı Tahmini: Karşılaştırmalı Bir Analiz

Karamollaoğlu H., Yücedağ İ., DOĞRU İ. A.

6th International Conference on Computer Science and Engineering, UBMK 2021, Ankara, Türkiye, 15 - 17 Eylül 2021, ss.139-144 identifier

  • Yayın Türü: Bildiri / Tam Metin Bildiri
  • Doi Numarası: 10.1109/ubmk52708.2021.9558876
  • Basıldığı Şehir: Ankara
  • Basıldığı Ülke: Türkiye
  • Sayfa Sayıları: ss.139-144
  • Anahtar Kelimeler: Customer churn analysis, Machine learning, Telecommunication
  • Gazi Üniversitesi Adresli: Evet

Özet

© 2021 IEEECustomer churn analysis is the process of predicting customers who tend to cancel the service (subscription) they receive for various reasons, especially in sectors such as telecommunications, finance and insurance, and determining the necessary operational steps to prevent this cancellation. The study used two separate datasets from kaggle.com to identify customers who tend to unsubscribe in the telecommunications industry. The analysis process was carried out by applying machine learning methods such as Logistic Regression, K-Nearest Neighbor, Decision Trees, Random Forest, Support Vector Machines, AdaBoost, Multi-Layer Sensors and Naive Bayes methods on the relevant datasets. It was seen that the most successful method in the customer loss analysis performed on both datasets was the Random Forest method.