7th Estonian Winter School in Computer Science (EWSCS)
VII Eesti Arvutiteaduse Talvekool (EATTK)
Palmse, Estonia
March 3 - 8, 2002
Prof. Heikki Mannila
Basic Research Unit,
Helsinki Institute of Information Technology
Helsinki University of Technology
Combinatorial problems in data mining
Abstract
Data mining aims at the development of efficient methods for the analysis of large data sets. The area combines methods from algorithms, statistics, and databases. We consider the combinatorial questions arising from the analysis of large 0-1 datasets and long sequences of events. Topics include
- approximate query answering using inclusion-exclusion and generalizations of Bonferroni's inequality
- combinations of combinatorial and probabilistic reasoning using frequent sets and maximum entropy methods We also present some open problems.
Course materials
Slide handouts will be distributed directly by the lecturer.
About the Speaker
E-mail: Heikki.Mannila@hut.fi
URL:
http://www.cs.helsinki.fi/~mannila/