E-Retail Example--Final Report

A Web-Mining Scenario Using CRISP-DM

The greatest deviation from the original project plan is also an interesting lead for further data mining work. The original plan called for finding out how to have customers spend more time and see more pages on the site per visit.

As it turns out, having a happy customer is not simply a matter of keeping them online. Frequency distributions of time spent per session, split on whether the session resulted in a purchase, found that the session times for most sessions resulting in purchases fall between the session times for two clusters of nonpurchase sessions.

Now that this is known, the issue is to find out whether these customers who spend a long time on the site without purchasing are just browsing or simply can't find what they're looking for. The next step is to find out how to deliver what they're looking for in order to encourage purchases.