UTCS Colloquia - Rui Mao, Shenzhen University, "Big data abstraction: metric-space indexing as an example"

Contact Name: 
Dan Miranker
GDC 3.516
Oct 4, 2013 10:00am - 11:00am

Talk Audience: UTCS Faculty, Grads, Undergrads, Other Interested Parties

Host:  Daniel Miranker

Talk Abstract: Variety is one of the challenges of big data.  Comparing to domain-specific data management solutions, generic solutions are more cost-effective, and thus have been adopted by commercial data management systems.  Looking back into the history of data management systems, one can clearly see the trend from domain-specific to generic solutions. Metric-space solution abstracts various data types into metric space and various similarity functions into metric distance functions.  It only takes use of the triangle inequality of metric distance functions for indexing, mining and so on, and possess great potential to be a generic solution to big data analysis.  After years of effort, fruitful results have been achieved in metric-space indexing, forming a good start point for metric-space data management of big data.

Speaker Bio: Dr. Rui Mao is an associate professor of the College of Computer Science and Software Engineering in Shenzhen University, China, and is the Executive Vice Director of the National High Performance Computing Center at Shenzhen, the Guangdong Province Key Laboratory of Popular High Performance Computers and the Shenzhen City Key Laboratory of Service Computing and Applications. His research interest includes distance-based data analysis, generic big data indexing and mining, high performance computing, data mining, statistical methods and bioinformatics.  Dr. Mao received BS(1997) and MS(2000) in computer science from the University of Science and Technology of China, and MS(2006) in Statistics and Ph.D. (2007) in Computer Science from the University of Texas at Austin. After three years as Senior Member of Technical Staff at the Oracle USA Corporation, he joined Shenzhen University in 2010.  Dr. Mao has over 40 publications in internationally renowned journals and conferences.  His work on the pivot space model for metric-space indexing won the SISAP 2010 Best Paper award.