hadryan/decision-tree

A machine learning decision tree

decision-tree

This project implements a machine learning algorthim (a decsion tree) that finds patterns in pre-existing data in order to determine how to categorize newly inputed data. More specifically, this function starts by taking in training data, a target attribute, and a list of all possible attributes. First, the function uses the target attribute input (i.e color) in order to assign each item in the training dataset to its corresponding group based on its characterization of the target attribute (i.e one group for red, one for green, etc.). The algorthim also determines the proper classification of each group based on the most common classification amongst it (i.e red items are fruits and green ones are vegetables). These groups are used to create "branches" in a decsion tree, and once the program finishes recurring (reducing the number of unused attributes from the inputted list of attributes), the final stored classifications are outputted as a "leaves."