Exmaple

OMIS 661: Business Intelligence Lab 7 & Assignment 6

Due: March 28, 2011

40 Points

Lab Exercise 7-1: Using Data Mining to Find Prospective Customers

Objective:

In this exercise, you will use the Microsoft Decision Trees algorithm to discover the bike-buying patterns stored in a historical customer list. The results will be stored in a data mining model. You will then use accuracy or lift chart to test the accuracy of the model. Lastly, you will apply the model to a new list of prospective customers to find out which are the most likely to buy a bike.

1) Open BIDS from SSAS, then create a new SSAS project called Lab7_DataMining.

2) Set up the data source of DB661 from the ecommerce.cob.niu.edu server. Note use service account for the impersonation information.

3) In the Solution Explorer window, create a data source view of DB661 to include the following tables: BikeBuyersTrainingSet, BikeBuyersTestingSet, and BikeBuyerProspects.

The data you will use is stored in three tables: BikeBuyersTrainingSet, BikeBuyersTestingSet, and BikeBuyerProspects. You will use the BikeBuyersTrainingSet table to discover the buying patterns. You will use the BikeBuyersTestingSet to test the accuracy of the model, and lastly you will use the model to find out which prospects in BikeBuyersProspects to target.

4) Right-click the title bar of the table BikeBuyerTrainingSet and choose Explore Data. Note that the data consists of personal demographic information, and there is a BikeBuyer column that indicates whether each person listed bought a bike or not. Also note that table has 7,000 rows. You will use this table to create a mining model that will store the patterns discovered by the algorithm. Note that names and addresses are not included.

5) Right-click the title bar of the table BikeBuyerTestingSet and choose Explore Data. Note that the data is similar to what you saw in the BikeBuyerTrainingSet table. Also Note that the...