
54cd7c5edcc3bff1647667c3fe66c3f9.ppt
- Количество слайдов: 1
Grant Number: IDM-0325464 Institution of PI: University of Georgia PIs: Amit P. Sheth, Krzysztof J. Kochut, John A. Miller, I. Budak Arpinar Title: Sem. Dis: Discovering Complex Relationships in the Semantic Web Research Objectives: Significant Results: Discover relationships among large semantic meta-base represented in RDF(S) or OWL 1. A large-scale ontology (SWETO) is built in RDF/OWL and populated using automated meta-data extraction techniques. (SWETO is made available to researchers for evaluating their tools and techniques) Employ ranking techniques to provide users with the most interesting and relevant results 3. A context-based ranking approach is completed and tested 2. Naive algorithms to find semantic associations including depth-first search, randomwalking, Tarjan's algorithm based-solutions completed. Preliminary evaluation shows some promise 4. A schema-based indexing and discovery algorithm is under development 5. A semantic query language is under development Approach: 1. Build large scale ontologies in RDF/OWL using ontology based meta-data extraction techniques 2. Algorithms for discovering semantic associations and similarities in large RDF/OWL graphs Graphic: Sem. Dis System Architecture Configuration of Ranking Criteria (using Context–selection based on classes/relations) 3. Ranking and filtering semantic associations using context, relevance, and trust 4. Definition of a semantic query language 5. Scalable indexing and query processing based on data-semantics Broader Impact: Recent advancements in information processing enable large scale semantic annotation of data. The next frontier is to automatically identify complex relationships between entities in this semantically annotated data. We develop a system that returns actionable information (with the associated sources and supporting evidence) to a user or application Ranked Semantic Associations (Entities: Chee-Keng Yap Ravi Ramamoorthi)
54cd7c5edcc3bff1647667c3fe66c3f9.ppt