Knight's Microsoft® Business Intelligence 24-Hour Trainer: Leveraging Microsoft SQL Server® Integration, Analysis, and Reporting Services with Excel® and SharePoint®

168



LESSON 19 Data Mining

EXPLORING THE MODEL

With the structure now created, you’ll need to deploy it to the SSAS instance. Then you can open

the Target Customers.dmm file from the Solution Explorer to open the Mining Structure Editor. In

the Mining Structure tab you can add additional columns into the mining structure. In the Mining

Models tab (shown in Figure 19-5) you can create additional models, like one for Association Rules

(as) and Clustering (cl), by right-clicking in the background and selecting New Mining Model. You

can also set columns to be ignored the algorithm.

FIGURE 195

Once the new models are deployed, you can view them in the Mining Model Viewer tab (shown in

Figure 19-6). If you have multiple algorithms, you can select the type of mining model you want to

see from the Mining Model drop-down box. From the decision tree algorithm you can see why the

mining model made the decisions it made. As you look at the model, you can see that the biggest

reasons the mining model thinks a customer is good are on the far left, and as you branch deeper

and deeper, the number of your customers who fit each particular condition becomes smaller.

For example, in Figure 19-6 you can see all the way to the right that the customers who fit this con-

dition are 99 percent likely to buy a bike. These customers are between 39 and 53 years old, have

two children, live in the Pacific reg