Mahadev Satyanarayanan, Carnegie Group Professor of Computer Science at Carnegie Mellon University, addressed the concept of "interactive data exploration" in a talk he gave at Yahoo! Research on Thursday, March 15. The standing-room-only audience was well engaged as Satyanarayanan discussed the challenges of finding vaguely-specified data items in the growing world of complex, loosely-organized data. He cited several industry examples, including the example of medical image retrieval and analysis, given the large amount of data that is generated in the form of photographs, x-rays and scans. According to Satyanarayanan, when such a large amount of complex data exists, examination of the data in some depth is first required to uncover and extract relevant queries. He explained this concept as hypothesis-formation and hypothesis-validation, or "interactive data exploration."

Satyanarayanan introduced Diamond, an open-source software platform for interactive data exploration that was jointly developed by Intel Research and Carnegie Mellon University. Diamond reduces data volume through the "early discard" of irrelevant data, resulting in better scaling. Collaborations with Merck and the University of Pittsburgh School of Medicine may result in more advanced pharmaceutical research and diagnostic strategies for imaging cancer lesions, skin lesions, glaucoma and transplant pathology.