The web app is at:
The inspiration was:
The data is from:
- The World Bank - Indicators
- World Economic Forum - Global Competitiveness Index
- WIN/Gallup International - End of the Year Survey
See below for details and tech stack
Similar countries to your top choice are found using machine learning (affinity
clustering). See the notebook /code/python-explore/affinitypropagation.ipynb
Data explorer fits a best-fit linear regression line.
For Data details, see /data/README.md
/heroku
contains the code for the deployed web application
/code
contains code to munge and explore the data [working]
Briefly, this is a Python-Flask web application with a PostgreSQL database and bokeh-driven data explorer, all deployed to Heroku. The web page is styles with Bootstrap. Favicon is my own, created on Paper for iOS and converted with favicon.cc
Python tools
- heroku deployment
- pandas data manipulation
- flask web app
Graphics
- bokeh
- D3, via NVD3
- Shiny
Analysis
- R
- Python
Web
- WTF-Forms for interactive web forms
- Bootstrap
- Custom CSS, HTML tweaks (Safari console debugger was very useful!)
Database (deployed remotely to Heroku/AWS):
- Postgres
Misc
- virtual env
http://www.usnews.com/news/best-countries/articles/methodology
65 country attributes, grouped into:
- Adventure
- Citizenship
- Cultural Influence
- Entrepreneurship
- Heritage
- Movers
- Open for Business
- Power
- Quality of Life
More on some CPIA indicators:
Fantastic tutorial for deploying python/flask apps to Heroku (a bit out of date though)
HTML Table styles in flask/pandas:
- https://sarahleejane.github.io/learning/python/2015/08/09/simple-tables-in-webapps-using-flask-and-pandas-with-python.html
- https://www.smashingmagazine.com/2008/08/top-10-css-table-designs/#8-rounded-corner
Flask and forms