This node visualizes a set of rules that are represented as a table
as it is the case with the Association Rule Learner. The items used to form the
antecedent and consequent of the rule are considered boolean attributes (with the
value 'present' if they occur in the rule). For every rule a circular icon in drawn.
The outer border is subdivided into as many segments as there are items occurring in
all the rules of the rule set. If the item is present in the antecedent, the corresponding
segment is colored in green. The interior consists of a pie chart encoding the confidence of
the rule. The color of the pie segment encodes the item of the rule consequent which may
be determined by a Color Handler node. The x- and y-locations of a rule icon are determined
by the recall and lift, respectively while the support is represented by the area of the icon.
Hiliting
Right-clicking a rule shows a popup menu which allows for hilighting options.
Tooltip
Move the mouse over the rule icons to get the exact values for the confidence, support and lift as well as the represented rule if plain text.
Options
Views
- Rule View
- The rule view draws a rule icon (see description above) for each rule contained in the set at the input port. Zooming the view is done via the mouse wheel. Panning is accomplished by right-clicking and dragging. If the the right mouse button is released without having moved since the it was pressed and it is over an icon, a hilighting popup menu is shown. Left-clicking a rule icon brings it to front and selects it by drawing a red border. For all three axes (x-axis, y-axis and icon size in terms of area) there exist scaling values that can be adjusted via the three sliders to left left of the split view. This may be necessary when the rule icons overlap each other or are depicted to far apart.
Workflows
- No workflows found
Links
- No links available
Developers
You want to see the source code for this node? Click the following button and we’ll use our super-powers to find it for you.