December 2018
Intermediate to advanced
274 pages
7h 46m
English
The goal of the decision tree is to arrive at the optimal choice for the given problem. The final leaf node should be the best choice for the problem at hand. The algorithm behaves greedily and tries to come to the optimal choice in each decision it takes.
The whole problem is divided into multiple sub-problems, with each sub-problem branching out to other sub-problems. The subsets arrived are based on a parameter called purity. A node is said to be 100% pure when all decisions will lead to data belonging to the same class. It will be 100% impure when there is a possibility of splitting its subsets into categories. The goal of the algorithm is to reach 100% purity for each node in the tree.
The purity ...
Read now
Unlock full access