Fig. 9
From: Cobdock: an accurate and practical machine learning-based consensus blind docking method
The representation of Radviz visualization for selected features by Boruta. The x and y coordinates of the data points correspond to the projected locations of the data points inside a two-dimensional space. It facilitates the identification of data clusters, patterns, and trends. Also, when a data point is situated at the origin (0, 0, 0), it signifies that the contributions of the characteristics (variables) represented by the axes are comparatively equitable and impartial for that specific data point. The feature names are represented in the “PROGRAM_NAME”_“FEATURE_NAME” format