August 2019
Intermediate to advanced
342 pages
9h 35m
English
Decision trees, therefore, shape their learning process based on a tree structure. Starting from a root node, subsequent decisions branch into various branches of different depths.
In essence, the samples dataset is divided by the algorithm in an iterative way, based on the decisions that are taken at each node, thus giving rise to the various branches. Branches, on the other hand, represent nothing more than the various ways in which data can be classified, based on the possible choices made at the various decision nodes.
This iterative process of subdividing the dataset is determined by a predefined measure of the quality of the subdivision conditions. The most commonly used metrics for measuring the ...
Read now
Unlock full access