Hyperlocal Air Quality Prediction in East Bay Area, CA

Building a Machine Learning Model for Air Quality Predictions

The goal of this project is to build machine learning models to predict air quality per city-block in the City of Oakland and San Leandro based on previously measured pollutant concentrations, local meteorological conditions, and local sources of emissions such as industries, traffic intersection data, and automobile traffic on highways without having to rely on complex physical modeling. A blog post on this work is available here!

The following files and folders are included in this repository:

Final Report and Slides:

1. Final Report
2. Slide Deck

Progress Reports:

1. Project Proposal
2. Data Wrangling
3. Data Story
4. Statistical Data Analysis
5. Milestone Report
6. Machine Learning Report

Jupyter Notebooks:

1. Data Cleaning - National Emissions Inventory Data
2. Daymet Data - API Call
3. Traffic Data - Open Street Maps
4. Distance to Facilities
5. Exploratory Data Analysis
6. Statistical Data Analysis and Machine Learning
7. Visualizations