16 June 10:00 Fabian Bohnert: Spatial Processes for Recommender Systems

Fabian Bohnert is visiting HIIT from Monash University (Melbourne,
Australia) week 25. Fabian is a PhD candidate there, see http://www.csse.monash.edu.au/~fabianb/. He gives a guest talk on Tue 16 June at 10 in C222.

Spatial Processes for Recommender Systems

Spatial processes are typically used to analyse and predict geographic data. In this talk, we adapt such models to the task of predicting a user's interests or item ratings within recommender systems. We present the theoretical framework for a model based on Gaussian spatial processes, and discuss efficient algorithms for parameter estimation. In the second part of the talk, we present a real-world application scenario for our model: predicting a museum visitor's interests in exhibits. We discuss our results from an evaluation with a real-world dataset of time spans spent by museum visitors at exhibits (viewed as implicit ratings). In this application scenario, our model achieves a higher predictive accuracy than nearest-neighbour collaborative filters.

Last updated on 3 Jun 2009 by Visa Noronen - Page created on 16 Jun 2009 by Visa Noronen