Mauro Brunato and Roberto Battiti

A Location-Dependent Recommender System for the Web


DIT, Università di Trento --- via Sommarive 14,
I-38050 Povo (TN) --- ITALY

Email: brunato|battiti@dit.unitn.it


Abstract:

User's location in a wireless mobile environment is an important
information item. Knowledge of the
position can be added to the existing user profile in order to provide
efficient services.

We propose a web-based recommender system
that automatically adapts relevance parameters of the recommendable
items (in our case web links) through implicit collaboration with
users, taking into account their position both during parameter
estimation and during the list generation phase.
After an
initial tuning phase, a specific URL will be recommended to a user
in a given location in a way that considers where and how often it
was accessed by the previous users.
A new middleware layer, the location broker,
associates past user positions and selected links. These
data are used to develop a spatial usage pattern for each site.

We describe a data structure that permits scalability and
analyze the empirical computational complexity both on a simulated
scenario and in a real-world context in our province.