As data collection technologies advance, a plethora of data is available in almost all domains of our lives, including science, medicine, government, business and finance. Knowledge discovery or data mining, concerned with extracting and abstracting useful knowledge from such massive data, is becoming increasingly more important, for purposes such as offering better service or care, or obtaining a competitive advantage.
In this course we introduce some of the main topics of data mining, including: generating predictive models; clustering or segmenting events into coherent groups; discovering new patterns, anomalies, and trends; and other abstractions. In addition to notions of various knowledge types and algorithms for their discovery, we will also discuss issues such as practical usefulness of discovered knowledge, scalability of algorithms, and ability to handle noisy data.
Prerequisites: One of CS 605 (Introduction to Database Management Systems) and CS 609 (Principles of Artificial Intelligence).
Required Text:
Data Mining: Concepts and Techniques.
J. Han and M. Kamber. (To be published by) Morgan Kaufmann
in 2000. The text will be available
from Bookstore at Student Union in September,
with kind permission from the authors.
Recommended texts:
Machine Learning. Tom Mitchell. McGraw Hill, 1997.
Data Mining Techniques for Marketing,
Sales and Customer Support. Michael Berry & Gordon Linoff.
John Wiley & Sons, 1997.
For more information, contact Dr Guozhu Dong at 775-5113 or gdong@cs.wright.edu.