Abstract

How can semantic data and semantic technologies be lever- aged for personalization and recommendation services? In this paper, we present SERUM (Semantic Recommendations based on large unstruc- tured datasets), a news recommendation system that utilizes semantic technologies to collect implicit user behavior and to build semantic user models. These models, combined with large-scale semantic datasets, are then used to compute personalized news recommendations using graph- based algorithms. We introduce the building blocks of SERUM for the se- mantic data management, personalization and recommendation, with the main focus on the implicit user behavior collection. Therefore, our sys- tem uses RDFa to collect meaningful user behavior and a self-developed user behavior ontology (the User Behavior Ontology, in short UBO) to build semantic user behavior models. The main contribution of this work is the introduction of the UBO and the associated semantic user tracking and modeling process.

@incollection {springerlink:10.1007/978-3-642-28509-7_37,
   author = {Plumbaum, Till and Lommatzsch, Andreas and De Luca, Ernesto and Albayrak, Sahin},
   affiliation = {DAI Labor, Berlin Institute of Technology, Ernst-Reuter-Platz 7, Berlin, Germany},
   title = {SERUM: Collecting Semantic User Behavior for Improved News Recommendations},
   booktitle = {Advances in User Modeling},
   series = {Lecture Notes in Computer Science},
   editor = {Ardissono, Liliana and Kuflik, Tsvi},
   publisher = {Springer Berlin / Heidelberg},
   isbn = {978-3-642-28508-0},
   keyword = {Computer Science},
   pages = {402-405},
   volume = {7138},
   url = {http://dx.doi.org/10.1007/978-3-642-28509-7_37},
   note = {10.1007/978-3-642-28509-7_37},
   year = {2012}
}