Tuesday, 5/22/2007
2:00 PM - 3:00 PM
Level: Technical - Intermediate

Syndication on the Web has attracted a great amount of attention in recent years (e.g., RSS). As technologies have emerged, there has been a transition to more expressive syndication approaches; that is subscribers and publishers are provided more expressive means for describing their interests and published content, enabling more accurate dissemination. However, today’s state of the art syndication approaches still provide relatively weak expressive power from a modeling perspective (i.e., XML and RDF are inexpressive languages) and provide very little automated reasoning support. In this talk, we discuss an expressive approach for content syndication on the Web, using semantic technologies. Specifically, we present an syndication framework based on the Web Ontology Language that provides many advantages, including a rich semantics-based mechanism for expressing subscriptions and published content (allowing finer control for filtering) and automated reasoning for discovering subscription matches not possible using traditional syntactic syndication approaches. Additionally, we discuss the application of such a syndication system in the financial news domain. This illustrates the advantages and experiences of using semantic technologies for content syndication. Further, this demonstrates the practicality of such a system through empirical evaluations on a financial news dataset.

Christian Halaschek-Wiener is a Ph.D. candidate at the University of Maryland working under Dr. Jim Hendler. His current research interests include expressive content syndication on the Web. More specifically, he is interested in using Web ontologies to represent published content on the Web, enabling finer grained filtering of information through automated reasoning. He is additionally working on incremental reasoning techniques, making such a framework usable in practice. Christian is particularly interested in applying these technologies to the financial domain, allowing the integration of qualitative information into quantitative financial models. Previously, Christian received a Masters degree in Computer Science from the University of Georgia.