SocialAirbnb Engineeringabout 4 hours ago
Airbnb personalizes search with guest journey model
9 min read
Airbnb built a Transformer-based sequence model that encodes years of guest behavior to surface relevant listings. The model learns from browsing patterns across multiple sessions, such as a guest searching San Francisco over several days. This approach aims to show the right listings at the right time.
Level
Hype check
Tap to vote and see what everyone thinks.
Summary by ByteBrief
