Part of the problem specification stage of a KDD project is to become familiar with the database. You should understand the fields, values and the semantics of both before proceeding. The statistics node provides a simple method of displaying a summary of all fields.
Using the method described for the data source node, add a stats node to the workspace.
Note that the node also appears in the project tree, but it is placed in the KDD Tasks section as it has not yet been connected to a data source.
Link the data source node to the statistics node so that the data can flow along the link.
To do this, select the link mode
button from the toolbar. The cursor changes again to show a small 
symbol, indicating the first stage of link mode. Click on the data
source node to make this the first node in the link; the cursor then
changes to show a 
symbol, indicating the second stage of link mode. Click on the
statistics node to complete the link.
An alternative method of linking nodes is to right-click on the first node (in the example, this is the data source node) then select the link command from the context menu. As you have already selected the first node in the link, simply click once on the statistics node to complete the link.
The stream should now look like this:
Right-click on the last node in the stream (the statistics node in this example), then select the run command from the context menu.
The data will be loaded into the data source node from the specified file. During the loading process, a dialog displays the loading process's progress and the number of records that have been loaded so far.
The data then flows down the link into the statistics node. WITNESS Miner calculates the statistics (and displays a progress dialog again). As each part of the KDD stream is completed, the links turn red.
Finally, a statistics results dialog appears, displaying summary information about each field in the database. This dialog is illustrated below. The information contained in the statistics results dialog allows us to understand the different types of fields: continuous numeric, discrete numeric and categorical. For numeric fields we can see the range of each field and for categorical fields, the number of unique values.
The statistics dialog looks like this:
Close the dialog to return to the main workspace.