Analysis of Toronto Neighbourhoods using Machine Learning

This is a repository for IBM Data Science Capstone Project. All analysis was done by Jessica Uwoghiren.
The datasets and Jupyter notebook used in the project is also included in this repository.
The full report is also included in this repository. See link to blog post for a more concise report -

Introduction

In 2019, 35% of new Canadian immigrants chose to settle in the City of Toronto. The City has 140 neighbourhoods, so, as a new immigrant, a vital question to answer is “What neighbourhood do I settle in?”. The aim of this project is to group Toronto neighborhoods in order of desirability using Machine Learning and Data Visualization techniques. I performed my analysis using on the following criteria:

• Total number of Essential Venues in each neighbourhood
  
• Primary and Secondary Benchmarks: Primary benchmarks considered were Unemployment rate, Crime rate and COVID-19 rates while the Secondary benchmark was housing price for a one-bedroom apartment in each neighbourhood.
  

Methodology

Python Libraries

The libraries used on this project include:

• Pandas – For storing and manipulating structured data. Pandas functionality is built on NumPy
• Numpy – For multi-dimensional array, matrix data structures and, performing mathematical operations
• Geopandas – For storing spatial data coordinates and shape files
• Scikit learn – For Machine learning tasks
• Plotly Visualization Package – For all visualizations (including maps and graphs)
• Requests - to send HTTP requests easily
• Geopy – To retrieve location coordinates

Project Flowchart

The main steps for this project can be summarized in the flow chart below:

Final Results

k-means clustering was used to group the neighbourhoods to obtain a final Neighbourhood Desirability index Choropleth Map