Number of items: 1.
and Herbrich, Ralf
and Graepel, Thore Matchbox: Large Scale Online Bayesian Recommendations.
We present a probabilistic model for generating personalised recommendations of items to users of a web service. The Matchbox system makes use of content information in the form of user and item meta data in combination with col- laborative filtering information from previous user behavior in order to predict the value of an item for a user. Users and items are represented by feature vectors which are mapped into a low-dimensional ‘trait space’ in which similarity is measured in terms of inner products. The model can be trained from different types of feedback in order to learn user-item preferences. Here we present three alternatives: direct observation of an absolute rating each user gives to some items, observation of a binary preference (like/ don’t like) and observation of a set of ordinal ratings on a user- specific scale. Efficient inference is achieved by approxi- mate message passing involving a combination of Expecta- tion Propagation (EP) and Variational Message Passing. We also include a dynamics model which allows an item’s popu- larity, a user’s taste or a user’s personal rating scale to drift over time. By using Assumed-Density Filtering (ADF) for training, the model requires only a single pass through the training data. This is an on-line learning algorithm capable of incrementally taking account of new data so the system can immediately reflect the latest user preferences. We eval- uate the performance of the algorithm on the MovieLens and Netflix data sets consisting of approximately 1,000,000 and 100,000,000 ratings respectively. This demonstrates that training the model using the on-line ADF approach yields state-of-the-art performance with the option of improving performance further if computational resources are available by performing multiple EP passes over the training data.
About this site
This website has been set up for WWW2009 by Christopher Gutteridge of the University of Southampton, using our EPrints software.
We (Southampton EPrints Project) intend to preserve the files and HTML pages of this site for many years, however we will turn it into flat files for long term preservation. This means that at some point in the months after the conference the search, metadata-export, JSON interface, OAI etc. will be disabled as we "fossilize" the site. Please plan accordingly. Feel free to ask nicely for us to keep the dynamic site online longer if there's a rally good (or cool) use for it... [this has now happened, this site is now static]