公开(公告)号：US20180253637A1

公开(公告)日：2018-09-06

申请号：US15446870

申请日：2017-03-01

申请人： Microsoft Technology Licensing, LLC

发明人： Feng Zhu ,  Xinying Song ,  Chao Zhong ,  Shijing Fang ,  Ryan Bouchard ,  Valentine N. Fontama ,  Prabhdeep Singh ,  Jianfeng Gao ,  Li Deng

IPC分类号： G06N3/04 ,  G06N5/04 ,  H04L29/08

CPC分类号： G06N3/0445 ,  G06N3/0454 ,  G06N3/08 ,  H04L67/22

摘要： A method to predict churn includes obtaining static features representative of a customer of a service, obtaining time series features representative of the customers interaction with the service, using a deep neural network to process the static features, using a recurrent neural network to process the time series features; and combining outputs from the deep neural network and the recurrent neural network to predict likelihood of customer churn.