With the tremendous growth in data capture, researchers are recognizing the need for knowledge and learning frameworks that can handle distributed data storage systems as well provide the means for distributed data analysis. Applying data analysis to practical problems requires deployment of a system that offers tools for both centralized and distributed knowledge discovery. This system should target the heart of the research activity, the knowledge discovery process, not just the end result. It should provide a means for locating datasets, retrieving data, segmenting the data into testing and training sets, data mining, and the publishing of results into a searchable repository. To address these needs, the NCSA Automated Learning Group has developed the D2K - Data to Knowledge analysis framework.

In addition, ALG has developed a number of D2K-driven or linked applications for performing specialized data mining tasks, including methods for dealing with text and image data. Links to these tools and further information about them are provided below.