![]() |
|
||
Client-Side Event Processing for Personalized Web AdvertisementRoland Stühmer1, Darko Anicic1, Sinan Sen1, Jun Ma1, Kay-Uwe Schmidt2, and Nenad Stojanovic1 1FZI Research Center for Information Technology, Haid-und-Neu-Straße 10-14, 76131 Karlsruhe, Germany
2SAP AG, Research, Vincenz-Prießnitz-Straße 1, 76131 Karlsruhe
Abstract. The market for Web advertisement is continuously growing and correspondingly, the number of approaches that can be used for realizing Web advertisement are increasing. However, current approaches fail to generate very personalized ads for a current Web user that is visiting a particular Web content. They mainly try to develop a profile based on the content of that Web page or on a long-term user’s profile, by not taking into account current user’s preferences. We argue that by discovering a user’s interest from his current Web behavior we can support the process of ad generation, especially the relevance of an ad for the user. In this paper we present the conceptual architecture and implementation of such an approach. The approach is based on the extraction of simple events from the user interaction with a Web page and their combination in order to discover the user’s interests. We use semantic technologies in order to build such an interpretation out of many simple events. We present results from preliminary evaluation studies. The main contribution of the paper is a very efficient, semantic-based client-side architecture for generating and combining Web events. The architecture ensures the agility of the whole advertisement system, by complexly processing events on the client. In general, this work contributes to the realization of new, event-driven applications for the (Semantic) Web. Keywords: Complex Event Processing, User Profiling, Rich Internet Applications, Semantic Web Technologies, RDF, RDFa, event-condition-action, ECA, Web Advertisement LNCS 5871, p. 1069 ff. lncs@springer.com
|