Information Integration Research Group

Welcome to the web page of the Information Integration Research Group of the Intelligent Systems Division of the Information Sciences Institute (ISI). ISI is part of the University of Southern California, but it is located off of the main campus in sunny Marina del Rey, California.

Our research group is developing intelligent techniques to enable rapid and efficient information integration. The focus of our research has been on the technologies required for constructing distributed, integrated applications from online sources. This research includes:

  • Information Extraction: Machine learning techniques for extracting information from online sources.

  • Source Modeling: Constructing a semantic model of wrapped sources so that they can be automatically integrated with other sources.

  • Record Linkage: Learning how to align records across sources.

  • Data Integration: Generating plans to automatically integrate data across sources.

  • Plan Execution: Representing, defining, and efficiently executing integration plans in the Web environment.

  • Constraint-based Integration:Interactive constraint-based planning and integration for the Web environment.

    We are applying these techniques to a variety of application areas including:

  • Geospatial Data Integration A mediator-based approach to organizing and integrating the huge amount of geospatial data that is available online.

  • Biological Data Integration. Application of the techniques and tools we have developed for other types of data to the biological data sources available online.