Brazilian E-Commerce Public Dataset by Olist: https://www.kaggle.com/olistbr/brazilian-ecommerce
Decision tree, Random forest, and Gradient boosting are used in this hands-on project to predict customer satisfaction of the e-commerce dataset.
Tree-Based Model Brief Explanation : Decision trees are one of the supervised machine learning methods for classification and regression. In this hands-on project, I will concentrate on the decision tree classifier and its various types, which will provide a readable classification model. The tree node on the decision tree classifier represents an attribute test. The branches emerge from that node, with each branch representing one of the previous attribute’s possible values. The leaf grows from those branches, with each leaf corresponding to the class labels associated with the instance. The instances in the training set are directed from the decision tree’s root to the leaf.
Project Overview : The goal of this project is to compare three modeling processes: decision tree, random forest, and gradient boosting to determine the most accurate and efficient machine learning modeling to predict customer satisfaction on an e-commerce dataset.