Systems Presentation

Mining Knowledge about Changes, Differences, and Trends

Guozhu Dong
Wright State University

Fri., Oct. 10th
3:30pm; 480 Dreese Labs
All interested parties are invited.
Pizza will be served after the talk.

Abstract:
Knowledge about changes, differences, and trends is very useful. For example,
companies wish to identify important temporal changes and trends in customer
purchase behavior, so that they can adjust their business priorities.
Medical researchers wish to identify differences in gene group interactions
between normal cell tissues and cancer cell tissues, so that they can identify
better treatment to cancer patients.

We discuss some recent results on mining such knowledge. We are concerned
with transactional data, relational data, and data cubes. We consider emerging
patterns that capture differences and changes between a dataset pair, gradient
patterns in a data cube that capture similar cells with big differences in
measure values, and multidimensional multi-level trends in sets of time series
in a data cube context. We discuss mining algorithms and ways to use the
patterns.

These talks are funded by a grant from the Ohio Board of Regents.