Topic outline
Introduction
Introduction to Data Mining and Knowledge Discovery
Regression Analysis
Multiple independent attributes
Dependent variable is Discrete/ Qualititative variable
Partition based Clusterings
Find 'k' numbers (user defined) of clusters with random seed selection.
- Focus on Advantages and Disadvantages
- Convergence criteria
- Method
Consider the attached e-commerce data. Cluster the customers and visualize them. Change the cluster numbers and comment on your observations.