Evaluation of customer behavior with temporal centrality metrics for churn prediction of prepaid contracts

作者：

Highlights：

• Evaluate interaction of customers in the telecommunication industry to predict churn.

• Modelize the interactions of customers with SNA using temporal and static graphs.

• Evaluate performance of different centrality metrics using two approaches.

• Apply similarity forests to evaluate the temporal time series.

摘要

•Evaluate interaction of customers in the telecommunication industry to predict churn.•Modelize the interactions of customers with SNA using temporal and static graphs.•Evaluate performance of different centrality metrics using two approaches.•Apply similarity forests to evaluate the temporal time series.

论文关键词：Churn prediction,Social network analysis,Similarity forests,Time series,Centrality metrics

论文评审过程：Received 11 December 2019, Revised 25 March 2020, Accepted 9 May 2020, Available online 10 June 2020, Version of Record 29 June 2020.

论文官网地址：https://doi.org/10.1016/j.eswa.2020.113553