98%(45)44 out of 45 people found this document helpful
This preview shows page 1 - 2 out of 6 pages.
DAT 220 Module 6Milestone 3My analysis for the clustering graph honestly reflects as organized of an approach as possiblewith limitations to what I can accurately pick out and process. With so many variables and possiblylimited user knowledge it can be difficult to identify what the cluster is actually depicting. With thelinear regression model and the logistical regression model, it was much easier to understand thedistribution of the data sets and to see patterns and trends. Before a successful run I had to selectvariables that were the most relevant to what I was looking to analyze. Otherwise, the graphsgenerated had too much info that skewed the data and created cluttered effect on the clusters andthe regression models. To keep things consistent, I chose to stay with Webstore Spending and Visitsversus age and income. The patterns that were discovered indicated that there was not a whole lot ofvariance between income and age when looking at the linear regression model for webstore