Decision tree classifier implementation using TDIDT (Top-Down Induction of Decision Trees) algorithm based on information gain heuristic, for continuous attributes.
Trained the classifier on Kaggle's Breast cancer dataset, and achieved 99.5% accuracy on training data and 94.7% accuracy on test data.
The resulting plot of the decision tree trained on the above dataset for maximum depth of 5: