Data: https://drive.google.com/file/d/1Wur14oUeOin27dBC_…
- Build a process that creates a set of association rules for this dataset, using a minimum support of 0.04 and a minimum confidence of 0.1. Show a screenshot of the Process panel.
- Run the process, and sort the association rules from highest support to lowest support. Show a screenshot of the list of rules. (You do not need to show all of them, but at least the first 10 must be visible.)
- What is the rule with the highest confidence? How would you interpret the confidence of that rule?
- What is the lift ratio (“lift”) of the rule “if FootballTalk, BaseballNews, then Investorettes”? How would you interpret the lift ratio of this rule?
- A few of the rules in the list have lift ratios that are less than 1. For instance, “if TVWatcher, then Investorettes” has a lift ratio of 0.701. What does that tell you about the relationship between those two Twitter accounts?
0 comments