adrientalbot / your-own-project

Welcome to Your Own Project!

Project Goals

• Do the e2e analysis by your own.
• Give some insights about a topic of your choice giving some arguments by:
  • Being able to transmit the analysis in an ordered way.
  • Using the statistical techniques learnt during the last weeks.
  • Plotting the graphs in a useful way for the reader / listener. Yes, no additional rules Jordi.
• Present your analysis in two ways and for two different audiencies:
  • Technical audience: you will present a paper.
  • Non tech audience: you will do a short presentation.

Deliverables

• A notebook with your analysis in a paper way. Jupyter or Zeppelin are allowed.
• A 3' presentation in the auditorium.
• Repository with your workflow + documentation + code. Even if you are working alone, you need to keep good practices!

The presentation

• In the auditorium.
• 3 minutes talk, 2 minutes questions. No more. Yeah, exactly.
• The audience will evaluate their understanding on your presentation (this will help you in further presentations).

General schedule

Today

• Think on a topic and propose some questions.
• Choose the data interesting.
• Look for documentation to give context to your project.
• Write the README file in your repository.

Tuesday

• Teachers validation --NO CODE UNTIL HERE--

Tuesday-Wednesday

• Start importing the data and cleaning it.
• Start the analysis. Remember all the techniques you have learnt.
• Prepare a draft of your first slides presentation (no analysis or conclusions yet): title, motivation, context, ...

Thursday

• First rehearshall. Take the feedback and use it!

Thursday - Next Tuesday

• Finish the analysis. Finish the slides.

Tuesday

• General rehearshall.

Tuesday - Wednesday

• Last improvements.

Wednesday

• Presentation!

General advices

• Organize yourself (don't get lost!).
• Ask for help vs Google is your friend.
• Define a simple approach first. You never know how the data can betray you ;)
• Documentate yourself. Know about the problem. Look for some analysis that someone did before you.
• You can use the projects your partners did in the last week. You can't use your own project.
• Before doing a graph, think what you want to represent.

Some inspiration

Explore the Google Trends

Hans Roslings's vizs

See an analysis about heart diseases

Your body's data