CS 5083 Knowledge Discovery and Data Mining

In today's information-rich world, finding automatic ways to sift through the wealth of information and to identify the salient patterns is critical. The process of knowledge discovery is about making sense of data that is too complex for humans to understand. Knowledge discovery is a interactive process between a human and a computer. Data mining techniques highlight salient patterns in the data. These patterns are examined by the human to provide feedback. Data mining refines the patterns and the cycle continues until the human is satisfied.

By the end of the semester, you will be able to:

• Recognize a knowledge discovery task
• Choose an appropriate knowledge representation (static vs dynamic, discrete vs continuous, propositional vs relational)
• Identify the elements of a successful knowledge discovery algorithm
• Apply current knowledge discovery algorithms to new situations
• Create new and successful knowledge discovery algorithms
• Evaluate and analyze the performance of new approaches

This will be a seminar style class where we will read and discuss current papers. In addition, we will have a class-wide project that we will work on and discuss.

Please note that most of the course material will go on Desire2Learn.

• Course syllabus
• Course schedule (updated as semester progresses)