Lehigh University

Dr. Tin Kam Ho

"Mirage: A Tool for Interactive Pattern Discovery, with Applications to the Virtual Observatory"

Friday, November 5, 3 PM

Packard Lab, Room 208

Abstract: The Virtual Observatory is a networked infrastructure for sharing massive and highly diverse observational data sets among astronomical researchers. To take full advantage of this vast resource, researchers need flexible data analysis tools that can handle huge amounts of data in a wide range of data types and meet a variety of demands for analysis and visualization, some at high speed. We describe the capabilities of Mirage, an experimental prototype of just such a versatile system for interactive pattern discovery (http://www.cs.bell-labs.com/who/tkh/mirage), and its applications within the Virtual Observatory. We focus especially on design issues: how should a data analysis tool be organized to support many different research goals and, furthermore, enable scaling up to even larger data sets and to address new problems. We identify several key open problems in data analysis and discuss fundamental technical challenges in data visualization and pattern recognition.

Bio: Tin Kam Ho is a Member of Technical Staff in the Computing Sciences Research Center at Bell Laboratories. Her interests embrace pattern recognition, data mining, and computational modeling and simulation. She received a Ph.D. in Computer Science from SUNY at Buffalo in 1992. She is Editor-in-chief of the journal Pattern Recognition Letters, and has served on the editorial board of several journals. In 1999 she received the ICDAR Young Scientist Award for her contributions to document image analysis and recognition. She is a Fellow of the IAPR, and has received 6 U.S. patents for her work in pattern recognition and image analysis. (tkh@research.bell-labs.com, http://www.cs.bell-labs.com/who/tkh)

		Dr. Tin Kam Ho "Mirage: A Tool for Interactive Pattern Discovery, with Applications to the Virtual Observatory" Friday, November 5, 3 PM Packard Lab, Room 208 Abstract: The Virtual Observatory is a networked infrastructure for sharing massive and highly diverse observational data sets among astronomical researchers. To take full advantage of this vast resource, researchers need flexible data analysis tools that can handle huge amounts of data in a wide range of data types and meet a variety of demands for analysis and visualization, some at high speed. We describe the capabilities of Mirage, an experimental prototype of just such a versatile system for interactive pattern discovery (http://www.cs.bell-labs.com/who/tkh/mirage), and its applications within the Virtual Observatory. We focus especially on design issues: how should a data analysis tool be organized to support many different research goals and, furthermore, enable scaling up to even larger data sets and to address new problems. We identify several key open problems in data analysis and discuss fundamental technical challenges in data visualization and pattern recognition. Bio: Tin Kam Ho is a Member of Technical Staff in the Computing Sciences Research Center at Bell Laboratories. Her interests embrace pattern recognition, data mining, and computational modeling and simulation. She received a Ph.D. in Computer Science from SUNY at Buffalo in 1992. She is Editor-in-chief of the journal Pattern Recognition Letters, and has served on the editorial board of several journals. In 1999 she received the ICDAR Young Scientist Award for her contributions to document image analysis and recognition. She is a Fellow of the IAPR, and has received 6 U.S. patents for her work in pattern recognition and image analysis. (tkh@research.bell-labs.com, http://www.cs.bell-labs.com/who/tkh)