DataScience Basics

1. Look into other useful datasets for this field.
2. Look into people who have done it before you.
3. Write the simplest stupidest solution, and see how it works.
4. Make some histograms and plots of the data.
5. Look at the data in binary format (using xxd or fancier tools).
6. Try to generate a synthetic similar-ish looking dataset, using procedural algorithms.