paper2:Real-time Personalization using Embeddings for Search Ranking at Airbnb

论文地址:

https://www.kdd.org/kdd2018/accepted-papers/view/real-time-personalization-using-embeddings-for-search-ranking-at-airbnb

1、论文简介

主要介绍了机器学习的嵌入技术在 Airbnb 爱彼迎房源搜索排序和实时个性化推荐中的实践

嵌入技术

(1)数据:用户点击会话

(2)训练方法:负抽样

(3)模型: Skip-gram model for Listing Embedding

(4)Booked Listing as Global Context,预定行为做长期特征

(5)User Type Embeddings 

(6)Rejections as Explicit Negatives

待整理。。。

paper2:Real-time Personalization using Embeddings for Search Ranking at Airbnb_第1张图片
listing embedding
paper2:Real-time Personalization using Embeddings for Search Ranking at Airbnb_第2张图片
user type embdding

参考资料:https://www.zhihu.com/collection/129459436

--2018.12.2

你可能感兴趣的:(paper2:Real-time Personalization using Embeddings for Search Ranking at Airbnb)