12th International Conference on Industrial Engineering , 2016-01-25

Title : ( Churn Prediction in Telecommunication Industry: A Data Mining Approach )

Authors: Mahnaz Sharafkhani , Hamidreza Koosha ,

Access to full-text not allowed by authors

Citation: BibTeX | EndNote

Abstract

Many industries use churn analysis as a means of reducing customer attrition, because it costs more to attract new customers than to retain existing ones. Telecommunication companies have started to use churn prediction with data mining techniques. In this paper, 3150 subscribers of one of the telecommunication operators were randomly selected. In the first part, the authors have utilized some binary classification techniques such as K-nearest neighbour, Naive Bayes, Decision Tree and Artificial Neural Network based on CRIPS-DM process. In the second part, clustering was performed with K-Means technique. The data is divided into three clusters. Classification techniques are done on each of these clusters separately. The results indicate that after clustering, decision tree with almost 95.5% of accuracy is the best technique among the mentioned classification techniques.

Keywords

Churn; telecommunication; data mining; classification
برای دانلود از شناسه و رمز عبور پرتال پویا استفاده کنید.

@inproceedings{paperid:1055502,
author = {Mahnaz Sharafkhani and Koosha, Hamidreza},
title = {Churn Prediction in Telecommunication Industry: A Data Mining Approach},
booktitle = {12th International Conference on Industrial Engineering},
year = {2016},
location = {تهران, IRAN},
keywords = {Churn; telecommunication; data mining; classification},
}

[Download]

%0 Conference Proceedings
%T Churn Prediction in Telecommunication Industry: A Data Mining Approach
%A Mahnaz Sharafkhani
%A Koosha, Hamidreza
%J 12th International Conference on Industrial Engineering
%D 2016

[Download]