Using the Data Mining Add-In for Microsoft Office - Microsoft Business Intelligence Tools for Excel Analysts - page 274

Figure 14-10: Results of the Forecast report.
When the tool completes, it appends the new predictions to the end of the source data table and
highlights the new values. The tool also creates a new worksheet named Forecasting Report. This
worksheet reports whether the wizard successfully created a prediction. The new worksheet also con-
tains a line graph that shows the historical trends along with the predicted values that were added.
The historical values are shown as a solid line and the predictions are shown as a dotted line.
Highlight Exceptions
Because “garbage in = garbage out” applies to data mining as much as with any data-oriented pro-
cess, a specific tool has been provided to help identify the garbage or outliers in your data. This is the
Highlight Exceptions tool, which relies on the Microsoft Clustering algorithm to identify rows with
similar characteristics.
The Highlight Exceptions tool analyzes and highlights data rows that do not appear to match the
general patterns found in the rest of the table. This not only highlights exception rows, it also high-
lights the specific column value in each row that is likely to be the cause of the exception.
To see the Highlight Exceptions tool in action, click the Highlight Exceptions button on the Analyze
tab. In the Highlight Exceptions dialog box, select the columns you want to evaluate to create your
distinct categories and click the Run button. You get a report similar to the one shown in
Figure 14-11.