A sunburst visualization of a BigML decision tree built on a dataset describing Titanic passengers and whether they survived.
The model uses BigML's upcoming automatic text processing to discover that passenger name honorifics (like "Mr.") are helpful for predicting survival.
The initial center circle represents the root of the tree. Each outer circle contains the children of the inner circle's nodes. The number of training instances captured by a node determine its arc length (or its size in radians).
Clicking on a node will zoom in to the subtree. After zooming in, selecting the new center point will zoom out one level.