Databases Reference
In-Depth Information
computer named “Argus” that has made the connection. It reported two
available displays.
After completing a VisMiner session, close the Control Center, which will
notify all connected slaves and visualizations that they too need to shut down.
Opening a dataset
VisMiner is designed to open datasets saved in comma-delimited text files (.csv)
and in Microsoft Access tables (MDB files). If your data is not in one of these
formats, there are many programs and “Save as” options of programs that will
quickly convert your data. Let's begin by opening the file Iris.csv, which is
contained in the data packet accompanying VisMiner. To open:
Click on the “File open” icon located on the bar of the “Datasets and
Models” pane.
Complete the “Open File” dialog in the same way that you would for other
Windows applications by locating the Iris.csv file. Note: If you do not see
any .csv files in the folder where you expect them to be located, then you
probably need to change the file type option in the “Open File” dialog.
Viewing summary statistics
All currently open datasets are depicted by the file icon in the “Datasets and
Models” pane. Start your initial exploration by reviewing the summary infor-
mation of the Iris dataset. To see summary information:
Right-click on the Iris dataset icon.
Select “View Summary Statistics” from the context menu that opens.
The summary for the Iris dataset (Figure 2.6) gives us an overview of its
contents. In the summary, we see that there are 150 rows (observations) in the
dataset, and five columns (attributes). Four of the five attributes are numeric:
PetalLength, PetalWidth, SepalLength, and SepalWidth. There is just one
nominal attribute: Variety. For each numeric attribute, the summary reports
the range (minimum and maximum values), the mean, and the standard
deviation. Nominal attributes have cardinality. The cardinality of Variety is
3, meaning that there are three unique values in the dataset. You can see what
those values are by hovering over the cell in the Cardinality column at the
Variety row. As you hover, the three values listed are “Setosa”, “Versicolor”, and
“Virginica”. The number in parentheses following the value indicates howmany
observations there are containing that value. In the Iris dataset there are 50
observations of variety Setosa, 50 of Versicolor, and 50 of Virginica.
