Dynamic behavior based churn prediction in mobile telecom

作者：

Highlights：

• Novel ways for modeling customer’s daily behavior.

• LSTM outperforms CNN in terms of learning customer's behavior representation.

• LSTM is efficient technique for predicting churn.

• It is important to consider the daily behavior of customer when predicting churn.

摘要

•Novel ways for modeling customer’s daily behavior.•LSTM outperforms CNN in terms of learning customer's behavior representation.•LSTM is efficient technique for predicting churn.•It is important to consider the daily behavior of customer when predicting churn.

论文关键词：Churn prediction,Mobile telecom,Machine learning,Deep learning,RFM,Dynamic behavior

论文评审过程：Received 25 January 2020, Revised 24 June 2020, Accepted 17 July 2020, Available online 23 July 2020, Version of Record 31 July 2020.

论文官网地址：https://doi.org/10.1016/j.eswa.2020.113779