Leveraging User Modeling on the Social Web with Linked Data

Abstract

Social Web applications such as Twitter and Flickr are widely used services that generate large volumes of usage data. The challenge of modeling the use and the users of such Social Web services based on their data has received a lot of attention in recent years. In this paper, we go a step further and investigate how the Linked Open Data (LOD) cloud can be leveraged as additional knowledge source in user modeling processes that exploit user data from the Social Web. Specifically, we introduce a user modeling framework that utilizes semantic background knowledge from LOD and evaluate it in the area of point of interest (POI) recommendations. For this purpose, we infer user preferences in POIs based on the users' behavior observed on Twitter and Flickr, combined with referable evidence from the Web of Data. We compare strategies that aggregate knowledge from two LOD sources: GeoNames and DBpedia. The evaluation validates the advantages of our approach; we show that the user modeling quality improves when LOD-based background information is included in the process.