Bartering Books to Beers:
a Recommender System for Exchange Platforms
Jérémie Rappaz† Maria-Luiza Vladarean† Julian McAuley‡ Michele Catasta†
† ‡
Bartering Books to Beers: a Recommender System for Exchange - - PowerPoint PPT Presentation
Bartering Books to Beers: a Recommender System for Exchange Platforms Jrmie Rappaz Maria-Luiza Vladarean Julian McAuley Michele Catasta What is barter? Def. Barter is a system of exchange where goods or services
a Recommender System for Exchange Platforms
Jérémie Rappaz† Maria-Luiza Vladarean† Julian McAuley‡ Michele Catasta†
† ‡
services are directly exchanged for other goods or services without using a medium of exchange.
First written mention of barter around 100’000 B.C
Need of a double coincidence No common measure of value
Solution: Online bartering platforms are specialized 20’000 daily visitors 250’000 registered users 18’000 registered users
Potential transactions Tom’s wish list
Heineken Duvel Brewdog Duvel Chimay Heineken Tom Jack Jane Rick
Jack’s wish list Jane’s wish list Rick’s wish list
Typical setting
Solution: Matching? 85K active users 2M items
Solution: Matching? 85K active users 2M items
have at least one swapping partner
Positive signals: wish-list + past transactions
ˆ yuj,ul,ik = pT
uj qik +
Users Items
n x m n x k k x m
Unidirectional interest
Make recommendations for one user but take into account reciprocal interest
ˆ yuj,im,ul,ik = f(ˆ yujulim, ˆ yulujik) = 1 2(ˆ yujulim + ˆ yulujik)
models a bias from one user to another.
ˆ yuj,ul,ik = pT
uj qik + social bias
z }| { sujul +
users’ behavi S ∈ R|U|×|U|
++ + Some pairs of users perform recurring trades.
Discard users/items that have been inactive for a long period
2005 2006 2008 2009 2010 2012 1 2 3 4 5
ˆ yuj,ul,ik = pT
uj qik + social bias
z }| { sujul + τuj δ(t; ¯ tuj ) + τikδ(t; ¯ tik) | {z }
temporal dynamics
Observed trade Negative sample
Bayesian Personalized Ranking (BPR)
see Rendle 2009
Maximizes AUC with positive examples only
MF MF+B MF+B+S MF+B+T MF+ALL
Bookmooch 0.758 0.798 0.849 0.938 0.958 Gameswap 0.790 0.842 0.863 0.890 0.903 Ratebeer 0.824 0.892 0.962 0.969 0.983
T S B
Bidirectionality Social bias Temporal dynamics
recommendation.
platforms.
temporal information.
Powered by SIGIR Travel Grant