Talk at 2nd Workshop on New Trends in Content-Based Recommender Systems, RecSys 2015 Vienna #recsys2015 #CoCoS
Authors: Lucas Bernardi, Jaap Kamps, Julia Kiseleva and Melanie Mueller. Booking.com, Amsterdam University, Eindhoven University.
Booking.com Travel recommendations address continuous cold start problem
1. Booking.com
The Continuous
Cold Start Problem
in e-Commerce
Recommender Systems
RecSys Vienna 2015-09-20
Lucas Bernardi, Melanie Mueller
, Amsterdam
Jaap Kamps, Julia Kiseleva
Amsterdam & Eindhoven University
7. Booking.com
• Cold start = not enough information (yet)
Classical Cold Start
→ Bridge period until warmed up
Other information:
popularity, content...
Standard
recommender
time →
performance
9. Booking.com
• New users:
- Millions unique users/day, growing
- Most not logged-in, no cookie
Cold Users
1 51 101 151 201 251 301 351
Day
ActivityLevel
Continuously cold users at Booking.com. Activity levels of two randomly chosen users
ver time. The top user exhibits only rare activity throughout a year, and the bottom use
• Sparse users: holidays are rare events!
10. Booking.com
• Classical cold-start / sparsity
new / rare users
User Continuous Cold Start
• Volatility
user interest changes over time
• Personas
• Identity
12. Booking.com
• Classical cold-start / sparsity
new / rare users
User Continuous Cold Start
• Volatility
user interest changes over time
• Personas
different interest at close-by points in time
• Identity
13. Booking.com
• Leisure versus business:
User Personas
1 11 21 31 41 51 61 71 81 91
Day
ActivtyLevel
Leisure
booking
Business
booking
inuously cold users at Booking.com. Activity levels of two randomly chosen
me. The top user exhibits only rare activity throughout a year, and the botto
nas, making a leisure and a business booking, without much activity inbetween
• Browsing on a sunny versus rainy day
• Weekend city trip versus long holiday
14. Booking.com
• Classical cold-start / sparsity
new / rare users
User Continuous Cold Start
• Volatility
user interest changes over time
• Personas
different interest at close-by points in time
• Identity
failure to match data from same user
16. Booking.com
• Classical cold-start / sparsity
new / rare items
Item Continuous Cold Start
• Volatility
item properties/values change over time
• Personas
item appeals to different types of users
• Identity
failure to match data from same item
27. Booking.com
Endorsements
• From users who stayed at hotel in destination
• Free text endorsements since 2013
• 2014: used NLP techniques to extract 256 tags
• 13 000 unique endorsements
• 60 000 destinations
32. Booking.com
- Implicit ratings:
buys, clicks, views...
Addressing Continuous Cold Start
- Content
- Ask user
→ surprisingly hard to beat!
→ not personalized
- item descriptions
- user profiles
- context
• Continuous cold start = information continuously missing
→ Need more/other information:
- Popularity
→ can add friction
→ weak signal
→ mix ‘em up!
Not only bridge warm-up period,
but deal with continuous cold start
33. Booking.com
• Classical cold-start
new & rare users / items
Summary: Continuous Cold Start
• Volatility
users/items change over time
• Personas
different interest at close-by times
• Identity
failure to match data from same user/item