Databases Reference
In-Depth Information
bathroom value, you may want to eliminate those rows with missing
values. However, you may also consider using other columns to predict a
value. A logical choice of predictors might be the number of bedrooms and
the area (square feet) of the home.
Check the “Predict value” radio button. A checked list box appears,
allowing you to specify which column values to use as input in predicting
missing bathroom values.
Check the “bedrooms” and “sqFeet” boxes.
Click “Next” to advance to the next column.
Predictions of numeric column values in VisMiner are accomplished by
building a linear regression model using the checked columns as input variables
and the column containing missing values as the output variable. Only those
observations containing values for all indicated columns are used to build the
model. Once built, the model is then used to generate estimated values for
the observations with missing values in the output (predicted) column. To
predict missing nominal (text) values, a decision tree is constructed and applied
in a similar manner.
Select the “bedrooms” column in the list on the left. For simplicity's sake,
select “Remove rows w/ missing value” as the handling option.
Select the “cul-de-sac” column in the list on the left, then select “Assign
default value” as the handling option.
In the “Default Value” box enter “N”.
Specify handling options for
the remaining columns according to
Table 3.1.
Click “OK”.
Once you have specified the handling option for each of the columns with
missing values and pressed “OK”, a new dataset named CmpltHomes.csv is
automatically created and saved in the same folder or database from which the
original was loaded.
Right-click on the original dataset, Homes.csv, then select “Close dataset”.
Right-click in the open space in the “Datasets and Models” pane, then
select “Reorganize dataset layout”.
View the summary statistics for the newly created dataset. Verify that all
columns with missing values have been handled.
Close the summary statistics window.
