What is KDD? Knowledge discovery in databases is the process of identifying valid, novel, potentially useful, and ultimately understandable structure in data.

• Structure » { pattern or model}: a parsimonious statement about subset of the data.

 

• The identified “knowledge” is typically used to
• Make predictions about new data
• Explain existing data
• Summarize a large database to facilitate decision making
• “Visualize” a high dimensional massive data set via extracting simplified local structure (as opposed to traditional global visualization techniques)

---