A Data Science Journey

MAP

Curriculum

Twelve tracks from software engineer to data scientist — with live progress.

01

Algorithms

Python solutions, explicit tradeoffs, and the patterns behind the problems.

06

Data Wrangling

NumPy arrays first, then pandas — messy data in, analysis-ready tables out.

05

Statistics

Distributions, uncertainty, and the summaries that make raw data legible.

VID

Videos

Courses worth the time, with the useful parts kept close.

Data Sourcing

Start with the data you can legally use: proprietary data your organization controls, public data open to everyone, or data purchased under license.

Some resources:

U.S. national government

Start with federal datasets published for public use.

U.S. state government

State portals expose regional records that federal datasets often miss.

European

Use the European Union portal for datasets published across Europe.

Non-Profit

Nonprofits publish focused datasets on health, development, and public welfare.

Private organizations

Private organizations release research data and specialized APIs.

Large datasets

These catalogs are built for datasets too large or broad for a single portal.

Web Scraping and APIs

When no dataset exists, collect structured data through an API or scrape it from the web with the right tool:

  • import.io
  • ScraperWiki
  • Tabular
  • Google Sheets
  • Excel

Google Sheets can import an HTML table directly. Put this formula in cell A1:

=IMPORTHTML('https://en.wikipedia.org/wiki/Iron_Chef_America', 'table', 2)