H. V. Jagadish
AT&T Labs
Florham Park, NJ
Abstract
I will present a broad overview of the Dr Know project (Data Reduction
and KNOwledge extraction for data Warehouses) at AT&T Labs. The focus
of this project is to enable data analysis on massive collections of data.
The central concept is reducing the large volume of data, through techniques
such as aggregation, sampling, and lossy compression.
I will try to suggest a unifying structure across many different data
representations, and present a couple of specific algorithms in somewhat
greater detail.
Luis Gravano
gravano@cs.columbia.edu