Welcome to DS2000 CS Practicum 8!
There are only two points on the agenda today:
You will get your midterms, let us know about any questions that you found difficult and we'll solve them together.
Start formalizing your project ideas and putting them down in this document: project_proposal_worksheet.docx. Submit this document as a team at the end of the practicum.
Start with what larger topic you are interested in as a group.
Think what questions/problems you want to address within that area.
While it doesn't need to be a strictly "novel research" idea, think about why you want to pursue it. "Because it's cool", is a perfectly good answer, you'll just have to frame it in a convincing way.
Think big, we'll narrow it down a bit later. Right now your idea might seem unattainable, but as you start programming on the project you will learn more and faster than through exercises.
Choose the datasets at the end, not at the beginning.
Use two or more datasets for really interesting results. For example: weather data on its own is not particularly interesting, taxi data and bike share data have been explored a lot, but what if we combine the three datasets and investigate the effects of weather on the choice of transportation mode?