MINIMUM: MergINg logIcal scheMas Using Mapping constraints

Integrating data sources on the level of metadata (i.e., schemas) is a long standing problem in databases. The application domains range from database design, data integration, to schema evolution. In this work, we aim at merging logical schemas, relational or nested relational, interrelated via schema mappings in a logical mapping language such as tuple-generating dependencies.

The goal of the project is to investigate various fundamental aspects of schema merging including:

• semantics: information capacity, logical  characterization of merge requirements
  
• functional/operational properties: query capability, updatability, incremental maintainability 
• algorithmic aspects:  feasible algorithms for schema merging
  
• expressiveness and efficiency: mapping languages  and complexities
  

Prototype:

We have developed a prototype MINIMUM capable of merging relational schemas using a finitely chaseable set of data dependencies in the format of tuple-generating dependencies and equality generating dependencies. The backend is implemented using Java and Prolog to perform reasoning using the chase procedure, while the frontend is developed using Eclipse RCP to present a Graphical User Interface.

Publications

2011

• Xiang Li, Christoph Quix

  Merging Relational Views: A Minimization Approach

  Published in Proc. of the 30th Int. Conference on Conceptual Modeling (ER 2011), 2011.

• Xiang Li, Christoph Quix, David Kensche, Sandra Geisler, Lisong Guo

  Automatic Mediated Schema Generation Through Reasoning Over Data Dependencies

  Published in Proceedings of the 27th International Conference on Data Engineering, ICDE 2011, April 11-16, 2011, Hannover, Germany.

2010

• Xiang Li

  Towards a Unified Framework for Schema Merging

  Published in VLDB 2010 PhD Workshop. Singapore, September 13-17, 2010.

• Xiang Li, Christoph Quix, David Kensche, Sandra Geisler

  Automatic Schema Merging Using Mapping Constraints Among Incomplete Sources

  Published in Proceedings of the 19th ACM international conference on Information and knowledge management (CIKM'10), October 26-30, 2010, Toronto, ON, Canada.