Guest lecture
Prof. Jeffrey D. Ullman
Stanford University
The MIDAS Project (Mining Data at Stanford)
Date Wednesday June 10th 1998 Place
Department of Computer Science
Teollisuuskatu 23 room A414 (4th floor)Time 12 - 14
Abstract
We discuss several different data-mining activities at Stanford. The general goal is to develop programmable tools for asking very broad questions about unstructed information, such as the Web or text repositories. Some new techniques have been used to build a better search engine, called Google. Other techniques have been used to search the Web for relationships that match examples, or to find pairs of words that have a correlated or causal relationship. Finally, we examine a tool for generalized association-rule mining. The ``query-flocks'' engine allows efficient compilation of broad data-mining queries that are not handled well by commercial SQL systems.
You are welcome
Heikki Mannila