gragh embedding相关论文小结（四）

关于一些工业界Graph Embedding论文的整理

• Billion-scale Commodity Embedding for E-commerce Recommendation in Alibaba
• Learning and Transfering IDs Representation in E-commerce
• Session-based Recommendation with Graph Neural Networks
• STAMP:Short-Term Attention/Memory Priority Model for Session-based Recommendation
• Neural Attentive Session-based Recommendation
• Real-time Personalization using Embeddings for Search Ranking at Airbnb

Billion-scale Commodity Embedding for E-commerce Recommendation in Alibaba
本篇论文基于Session-based Graph embedding来实现RS。而RS分为两阶段：match和rank，简单的说就是根据user或者query来选出candidate items set，之后给其中每一个item打分来排名发给user从而实现推荐，这篇论文关注的是math阶段的解决方法，考虑的是side info和item embedding。既然是Graph embedding方法，就得有Graph。这里是做推荐，采用的也是已有的Session-based方法，就是根据user交互过的items，在同一个sessi