[Back]


Scientific Reports:

H. Huang:
"Context-aware Collaborative Filtering in Location Based Services";
Report for Austrian Marshall Plan Foundation; 2010; 29 pages.



English abstract:
Technical advances in mobile devices and mobile communication have led to the introduction of Location Based Services (LBS). Currently, providing context-aware services/information is still a key challenge in LBS applications. Collaborative filtering (CF), known as Amazon-like recommendation, is a promising solution for providing context-aware recommendations. The goal of this report is to investigate how context-aware CF (CaCF) can be introduced into LBS to provide context-aware recommendations. Specifically, we focus on applying CaCF methods on the highly available spatio-temporal trajectories to enhance visitors with context-aware POI (Point of Interest) recommendations.
This report first proposes a two stage method to identify context parameters which are relevant and thus needed to be modeled in CaCF. After identifying relevant context parameters, we explore two different approaches (i.e., Local-Global Approach LGA, and Statistic-Based Approach SBA) to measure similarity between different contexts (situations). In considering two different ways of incorporating context information into the CF process, four CaCF methods are designed for LBS: LGA_CP_CaCF (using LGA and contextual pre-filtering), LGA_CM_CaCF (using LGA and contextual modeling), SBA_CP_CaCF (using SBA and contextual pre-filtering), and SBA_CM_CaCF (using SBA and contextual modelling). With these CaCF methods, smart services like "in similar context, other people similar to you often ..." can be provided.
Finally some experiments are designed to evaluate the proposed methods. The results of the experiments show that the proposed CaCF methods are feasible and useful for providing context-aware recommendation in LBS applications. Also, we prove that including context information in the CF process can improve the predictive performances in LBS.

Keywords:
context-aware computing, collaborative filtering, context similarity, context-aware recommendation, trajectory

Created from the Publication Database of the Vienna University of Technology.