Fig. 4
From: Discovering feature relevancy and dependency by kernel-guided probabilistic model-building evolution
The visual program implementing Kiedra in the Orange/ Goldenberry workbench. Widgets on the left–hand side of the canvas provide the required input (data sampling, kernel machine setup, etc.) for the central discoverer component (Kiedra) which executes the algorithm so as to estimate relevancy and dependency, and then provides these results as inputs for the collection and visualisation widgets on the right–hand side of the canvas. Two additional widgets shown on the top of the canvas were used to implement wKiera and to execute ReliefF. See text for further details