The group is currently working on six projects:
- Data Interoperability for Geospatial Applications This project integrates and extends Computer Science research in the Database Management System (DBMS) area with a State of Wisconsin Land Information System (WLIS). Web-based DBMS user interface, search, and query techniques will be incorporated in WLIS. The concept of WLIS as a statewide access mechanism for land related data in Wisconsin has been actively under development for several years by a working group that includes various levels of government and the University of Wisconsin. WLIS will be a distributed Web-based system with heterogeneous data residing on local and state servers. A large part of WLIS is spatial, Geographic Information Systems (GIS) data. Collaborating institutions: UIC (lead) and University of Wisconsin-Madison. (Paper)
- CASSIS: Context-Aware Computing with Applications to Public Health Management This project creates an architecture for a system that provides comprehensive support for context-aware applications. More than 130 publications have resulted from this project among collaborating institutions UIC (lead), Brown University and Northwestern University. The current research emphasis by the ADVIS group is on an extension to a role-based access control (RBAC) approach to provide for the dynamic association of roles with users in collaborative environments. (Paper)
- Semantic Desktop With the Semantic Web vision, a number of Personal Information Management Systems have been proposed [Web page for the 1st Workshop on the Semantic Desktop]. In our approach, we propose: (1) Semantic data organization based on multiple layers of ontologies; (2) Flexible data manipulation provided by data interoperation and supported by an expressive query language; and (3) Rich visual manipulation that uses the underlying semantic connections among data. (Paper)
- Efficient Matching for Large Real-World Schemas and Ontologies This project extends the state of the art in schema and ontology matching, and therefore in data integration, by testing and evaluating methods and strategies that establish relationships among semantically related concepts in heterogeneous data sources.
The following research issues are addressed: (1) Design of methods and algorithms for schema and ontology matching that operate at different levels of granularity (e.g., concept, structure); (2) Development of a prototype of an integrated system that supports the visualization and manipulation of large schemas and ontologies in addition to the developed matching methods and algorithms; (3) Test and evaluation of the above methods and system prototype in terms of their effectiveness, including accuracy (precision, recall) and efficiency (execution time). (Paper, Demo Paper, Video)
- Privacy-preserving record and image mapping In this project we consider methods to perform record and image mapping in a privacy preserving way, more specifically, we will develop mapping procedures with privacy guarantees for the parties involved in the mapping process. Quality measures will be developed and uncertainty will be considered. Collaborating institutions: UIC (lead), Brown University, Rutgers University, University of Rome "La Sapienza." (Preliminary paper)
- Query Languages for Heterogeneous Transportation Networks We are exploring textual and visual query languages for graphs that represent urban transportation networks. Issues to explore include the syntax and semantics of the languages, their expressive power and query evaluation. This project is part of an IGERT grant at UIC. (Paper)
- Visualization and Integration of Heterogeneous Information (DelaunayMM and DelaunayViewsystems)
This system tightly integrates a visual query language for distributed queries with the layout specification that can emphasize the information semantics. We are continuing this work in several directions, including the use of ontologies to aid in query refinement, the integration of security concerns, and the creation of metadata for the synchronized retrieval and presentation of the information.
- Visual Query Languages ( Delaunay system)
The basis of this system is a visual meta-query language for displaying and querying object-oriented databases, called DOODLE. Features of the system include the support of user defined query languages and visualizations. Currently we are continuing work on: (1) semi-automatic visualizations; (2) use of metadata to generate visualizations; and (3) using dynamic constraint solving for positioning optional visual components.
- PANDA (Publishing Assistant for Networked Document Assembly)
This project is on a powerful authoring tool that integrates and presents marked-up information that is available on the web and gives powerful customization capabilities to the users (publishers and end-users). The result of this research is being directly incorporated into the ACM SIGMOD Digital Symposium Collection. We are extending PANDA to incorporate: visual query languages and metadata support.
- CASA (Classification using a Structural Approach)
We have been investigating automatic classification of Web documents. We studied the recall based on the structural analysis of more than 600,000 Web documents. We are investigating new algorithms that improve recall.
- MATIC (Multimedia Authoring with Time Constraints)
This is a multimedia authoring project for the temporal synchronization of multimedia streams. We have devised an efficient algorithm that determines whether synchronization can be achieved. Future work includes integrating this approach into SMIL.
This project optimally partitions the software components of a typical algorithm animation system and leverages the power of the Java language. This system is currently used for geometric algorithms (with the Computational Geometry Center at Brown U.).
- WebSA (Web Site Agent)
In this project users upon entering a Web site are given hints reflecting the interests of the previous thousands of visits to the site. In order to take into account the information previously gathered, efficient database mining of the previously collected data and the building of relevant aggregate information are performed. We are developing other collaborative filtering algorithms within the same architecture.
- ACM DiSC
Until 2001 the lab was "home" to the ACM SIGMOD Digital Symposium Collection (DiSC), an electronic journal that is the most extensive yearly source of database publications.