Customer purchase prediction from the perspective of imbalanced data: A machine learning framework based on factorization machine
作者：
Highlights：
• A method to obtain accurate prediction result of customer purchase is proposed.
• A feature combination method is introduced to alleviate data sparsity problem.
• The maximum marginal category is used to distinguish the overlapping sample.
• The cost sensitive learning is embedded to obtain the imbalanced prediction results.
摘要
•A method to obtain accurate prediction result of customer purchase is proposed.•A feature combination method is introduced to alleviate data sparsity problem.•The maximum marginal category is used to distinguish the overlapping sample.•The cost sensitive learning is embedded to obtain the imbalanced prediction results.
论文关键词：Customer purchase prediction,Factorization machine,Imbalanced data,Machine learning
论文评审过程：Received 9 July 2020, Revised 17 January 2021, Accepted 16 February 2021, Available online 23 February 2021, Version of Record 5 March 2021.
论文官网地址：https://doi.org/10.1016/j.eswa.2021.114756
