1. Calculate the significance of each feature in data-splitting.
2. Then, pick the feature with the highest significance, and split the data by this. This means that your test will check which of the different values of the feature the example meets.
3. Repeat Step 2 under each node, and so on, until an answer is achieved.
Congratulations!
Congratulations, you have just learned the conceptual basis for Decision Trees! In the notebook coming up, we will go through the steps involved in implementing it in Python, but suffice it to say: you have finished the hard part!
Previous Section
5.3 Entropy and Information GainNext Section
5.5 ImplementationCopyright © 2021 Code 4 Tomorrow. All rights reserved.
The code in this course is licensed under the MIT License.
If you would like to use content from any of our courses, you must obtain our explicit written permission and provide credit. Please contact classes@code4tomorrow.org for inquiries.