( Unifying Search and Reasoning from the Viewpoint of Granularity )

 

USeR-G ( Unifying Search and Reasoning from the Viewpoint of Granularity ) is a set of strategies that aim at combining the idea of granularity and ReaSearch (Unifying Search and Reasoning) at Web Scale to solve the diversity and scalability problems. It emphasize to make incomeplete Web scale reasoning focus on appropriate levels and views on the web of data, neglecting irrelevant information, in order to find more important results in the time-bounded and Web-scale environment.

The name "USeR-G" for the set of strategies comes from two perspectives. Firstly, it is the abbrieviation of the umberela term for this study. Secondly, these methods are considered from the users perspective.
 
Main Strategies :
 
  • Multilevel Completeness Strategy,
  • Multilevel Specificity Strategy,
  • Multiperspective Strategy,
  • StartingPpoint Strategy,
  • Multilevel Certainty Strategy,
  • et al.
 
 
Released Source :
 
1. Computer Scientists Coauthor Network : We bulit the Computer Scientists Coauthor Network as an RDF file so that one can develop applications based on this network and utilize the relationships among the authors. Some case studies can be found in our publication [Download the Coauthor Network RDF file].
 
2. Removed Keywords for Building Domain Knowledge Structures : We built domain knowledge structures based on publication structures in a certain domain, in order to provide a good domain knowledge structure, we filtered some irrelevant words. As an illustrative example of building domain knowledge structures, we built a knowledge structure of "Artificial Intelligence" based on relevant proceedings and journals structures in DBLP [Download the filtered words].
 
3. The Construction of "Artificial Intelligence Ontology" : We build an "Artificial Intelligence Ontology" by ourselves since this ontology contains all the possible branch information that has ever been proposed as sections or subsections in AI related conferences in the DBLP dataset and is relatively complete. [Supporting Material]
 
Development Team :
 
  • Yi Zeng : Ph.D Candidate, International WIC Institute, Beijing University of Technology.
  • Yan Wang : Ph.D Candidate, International WIC Institute, Beijing University of Technology.
  • Peiqiang Li : Master Candidate, International WIC Institute, Beijing University of Technology.
 
 
Steering Committee :
 
  • Ning Zhong : Professor, Maebashi Institute of Technology, Japan.
  • Yulin Qin : Senior Research Scientist, Carnegie Mellon University, U.S.A.
  • Zhisheng Huang : Senior Researcher, Vrije University Amsterdam, The Netherlands.
  • Yiyu Yao : Professor, University of Regina, Canada.
 
 
Related Publication :
 
  • Unifying Web-scale Search and Reasoning from the Viewpoint of Granularity. Yi Zeng, Yan Wang, Zhisheng Huang, Ning Zhong. In: Proceedings of the 2009 International Conference on Active Media Technology, Lecture Notes in Computer Science, Springer [to appear]. (Related to various strategies of the USeR-G method)
  • The Quest of Parallel Semantic Web Reasoning. Peiqiang Li, Yi Zeng, Jacopo Urbani, Spyros Kotoulas, Ning Zhong. In: Proceedings of the 2009 International Conference on Active Media Technology, Lecture Notes in Computer Science, Springer [to appear]. (Related to initial thoughts on USeR-G parallelization)
  • DBLP-SSE: A DBLP Search Support Engine, Yi Zeng, Yiyu Yao, Ning Zhong. In: Proceedings of the 2009 IEEE/WIC/ACM International Conference on Web Intelligence, IEEE Computer Society, Milan, Italy, September 15-18, 2009 [to appear]. (Related to starting point strategy and search refinement)
 
 
Project Support :
This study is supported by the research grant from the European Union 7th framework project FP7-215535 Large-Scale Integrating Project LarKC (Large Knowledge Collider).
 
Copyright by International WIC Institute, Beijing University of Technology, P.R. China.