This is a project I did for my Machine Learning (CP 8318) class at Ryerson University. The goals of this project are:

1. To see if there are trends in the data between the types of cannabis, rating, effects, and flavour,
2. to develop algorithms that can aid in the selection of a cannabis strain based on the description.

The iPython file is given as well as my report on the project which explains in detail the methods and results. The dataset used is found on kaggle and named 'cannabis.csv'. Link: https://www.kaggle.com/kingburrito666/cannabis-strains. The dataset consists of 6 features: Strain (name of strain), Type (indica, sativa or hybrid), Rating (from 0-5), Effects, Flavour, Description.

The data is collected from internet users on the website Leafly. This type of data collection often involves biases due to the subjective nature of the experience of individual cannabis strains.

This project was done in Canada after October 17th, 2018. Recreational cannabis was legalized on this day, and this project was done entirely after this date. This project is not meant to be used as a medical advice for choosing cannabis strains, if you are thinking about using cannabis medically, please consult your doctor.