ITR: COLLABORATIVE RESEARCH: A Data Mining and Exploration Middleware for Grid and Distributed Computing

The purpose of this project is to develop and implement a middleware that allows mining and analysis of distributed data. The research will address an increasing interest in the exploration and mining of the large volume of data that is generated by the business, scientific, engineering, academic and defense communities.

Novel techniques will be developed in the following areas:

Data and Policy Management Services: These services will provide organizational level access restriction capabilities for the owners of the data while allowing users to access a more efficient transport of data.

Data Mining and Exploration Services: This framework will include a library of data mining tools that will work effectively for example if: data is distributed on multiple sites, the user has varying privileges depending on what site they are accessing, a user wants to trade off computing time versus accuracy or a user wants to access the site remotely.

Scheduling and Replication Services: System administration will be based upon policies at the participating organizations and their established privileges for users. Computation and data will be scheduled jointly to optimize metrics.

This combination of distributed computing and data mining will be made widely available to students, researchers, and other interested groups in government, industry and education.