Predicting interpurchase time in a retail environment using customer-product networks: An empirical study and evaluation
作者：
Highlights：
• We leverage similarities in customer purchases for product attrition prediction.
• We extract an extensive feature set from customer-product graphs.
• We boost predictive performance by 6% and product identification by 20%.
• Our model illustrates the importance of transactional data for marketing.
• We show the advantage of network-based analytics in offline retail.
摘要
•We leverage similarities in customer purchases for product attrition prediction.•We extract an extensive feature set from customer-product graphs.•We boost predictive performance by 6% and product identification by 20%.•Our model illustrates the importance of transactional data for marketing.•We show the advantage of network-based analytics in offline retail.
论文关键词：Customer-product graph,Interpurchase time,Offline retail,Purchase behavior,Social network analytics,Transactional data
论文评审过程：Received 11 November 2017, Revised 23 February 2018, Accepted 11 March 2018, Available online 12 March 2018, Version of Record 20 March 2018.
论文官网地址：https://doi.org/10.1016/j.eswa.2018.03.016
