This repository contains the final capstone project for IBM's Data Science Professional Certificate course on Coursera, completed in May-June 2019.
The capstone project focuses on analyzing and comparing neighborhoods using data science techniques, with a particular emphasis on clustering algorithms and location-based data analysis.
The primary capstone project that analyzes and compares neighborhoods to identify patterns and similarities. The project is structured in two phases:
- Week 1: Initial data collection, exploration, and problem definition
- Week 2: Advanced analysis, modeling, and final results
A comprehensive analysis of Toronto neighborhoods using machine learning clustering techniques, divided into three parts:
- Part 1: Data collection and preparation - gathering neighborhood data and setting up the analysis framework
- Part 2: Feature engineering and exploratory data analysis
- Part 3: Clustering implementation using K-means algorithm and visualization of results
- Python with Jupyter Notebooks
- Machine Learning: Clustering algorithms (K-means)
- Data Analysis: pandas, numpy
- Visualization: matplotlib, folium for mapping
- APIs: Location data and venue information (Foursquare API)
- Geospatial Analysis: Working with coordinates and mapping neighborhoods
- Multiple Jupyter notebooks documenting the entire data science workflow
- Comprehensive PDF report with findings and methodology
- Data visualizations and cluster analysis results
- Reproducible code for neighborhood segmentation
The project applies data science techniques to segment and analyze Toronto neighborhoods based on various features such as venue types, location data, and other characteristics. This analysis helps identify similar neighborhoods and provides insights into urban patterns and neighborhood characteristics.
This project demonstrates practical applications of data science in urban analysis and location-based intelligence.