Face to purchase: Predicting consumer choices with structured facial and behavioral traits embedding
作者：
Highlights：
• A hierarchical model that combines face embedding with structured behavioral traits embedding for purchase prediction.
• Feature engineering of structural behavioral traits and multi-faceted face features for generating a graph structure of images.
• Selective graph convolution based on Graph Convolutional Network and light Inception for leveraging graph information effectively.
摘要
•A hierarchical model that combines face embedding with structured behavioral traits embedding for purchase prediction.•Feature engineering of structural behavioral traits and multi-faceted face features for generating a graph structure of images.•Selective graph convolution based on Graph Convolutional Network and light Inception for leveraging graph information effectively.
论文关键词：Purchase prediction,Hierarchical embedding,Graphical neural networks,Correlation analysis
论文评审过程：Received 9 August 2021, Revised 25 October 2021, Accepted 27 October 2021, Available online 2 November 2021, Version of Record 11 November 2021.
论文官网地址：https://doi.org/10.1016/j.knosys.2021.107665
