A.11.11 Column Classification Window

Column Classification Window

The Column Classification Window takes a column or other algebraic expression, and generates a number of mutually exclusive Row Subsets based on its contents. For instance if you have a column with different integer values representing different object types, this window will let you add a new subset to the table for each of the distinct values (object types) appearing in the table. Each subset contains rows with a single value of the classification expression you supply. In the above example, a new subset will be created for each of the four most commonly-occurring constellations in the table.

You can activate this dialogue using the Classify By Column () button in the Subsets Window.

In the Query panel, you specify the expression you want to classify on. You can select a simple column name from the drop-down list, or type in an algebraic expression as described in Section 7. If you want to classify on ranges of values rather than exact equivalence you can use an expression that rounds to an integer, for instance "toInteger(RMAG/2.0)*2" would give you subsets corresponding to bins of width 2 in magnitude. When you have entered the expression, you may need to click the Classify button to start the classification (or it may happen automatically). The classification will often complete straight away, but for large tables it could take a noticeable amount of time, in which case a progress bar is shown at the bottom of the window. You can stop a long-running classification in progress with the Stop button. If you have a very large table with many distinct categories in the given column, the process can take a lot of memory - if the application runs out of memory, a warning will pop up and the classification will not complete.

When the classification has run, the results are displayed in the Results panel. Two fields control the way these results are displayed:

Number of Categories

Since in general there may be a large number of different values in the column of interest (as many as there are rows), only the few most popular ones are shown. This field controls the maximum number shown, you can adjust it as required. The fixed value after the entry field shows the total number of distinct values discovered in the data; increasing the field's value beyond this value will have no further effect.

Subset name prefix

Defines the prefix added to the column values to give the default name for each subset that will be added to the table. The application tries to come up with something sensible based on the classification expression, but doesn't always succeed. If you type in a new prefix here, all the subsets listed below are given new default names accordingly.

The results are displayed in a table, each row corresponding to a Row Subset that can be added to the table based on the classification. The subsets are shown in decreasing order of popularity (the subset containing the most rows is listed first). The final row, labelled "other", groups all the rows which are not in any of the other currently selected and displayed subsets. The following columns are shown:

Count

Number of rows in the subset.

Value

The classification value shared by all rows in the subset.

Subset Name

The name of the subset that will be added. The applicaiton tries to come up with sensible default names, but doesn't always succeed - you can edit this field as required. If the name matches the name of any subset already present in the table, when the new subset is added it will replace the old one.

Add Subset?

If this box is checked, the Add Subsets button below will add a subset corresponding to this row; if not, it will be ignored. If unchecked, the rows from this subset are considered part of the other subset listed at the bottom (you can see the Count field in the other row adjusting when you include or exclude subsets by checking/uncheking this box).

When you have adjusted subset names and selected the ones you want to add, click the Add Subsets button at the bottom, and one subset will be added to the table for each of the items with the Add Subset? checkbox ticked. If you don't want to add any of these subsets after all, just hit Cancel or close the window.