baltimoremaps
By: Adam Li
This is an interactive web app that utilizes python for data preprocessing/mining and d3.js for data visualization. It creates a map of the USA based on census of the different counties shape files and then populates colors based on # of republican vs. democrat votes within each state.
The tools required to get the data working is:
- ogr2ogr
- topojson
These are both downloadable via npm, which is the node package manager.
1. Data Cleaning
This is the first step in the project. Refer to the jupyter notebook for how the data is processed from .shp files into topojson files.
2. Data Analysis
This step is pretty simple. I just pull data from http://www.cnn.com/election/primaries/counties/ia/Dem, or Kaggle: https://www.kaggle.com/benhamner/2016-us-election and compute the % of republican voters within a state.
3. Data Visualization
This step uses d3.js, colorbrewer, jquery and queue to produce a chloropleth map of the USA based on % of republican voters with a colorrange from red to blue. D3.js also allows enhanced interactivity, where I added the ability to zoom in on a state and see some of the largest city locations within that state.
To Do:
In the future, this repository would be great to help get set up in a data science project that utilizes geodata and visualization of a map using d3.js.