Week 5

Let's start off with a quick warm-up. Click on the link below and try out the puzzle:

A Quick Puzzle to Test Your Problem Solving - The New York Times (nytimes.com)

Key Takeaways

1. Bias affects both humans (and machines). Did you discover any biases you have while doing this activity?

2. Avoid “confirmation bias” by looking for both “positive” and “negative” examples. Did you feel comfortable making a guess even if you thought it would be wrong?

3. The best predictions are made based on seeing diverse examples. If you only try a few guesses, your prediction might miss key details that would have been more apparent with a broad dataset. In same way, AI models that encounter diverse examples during "training" will also "learn" to perform better.

Watch this video with your team, then discuss amongst yourselves the question we ask at the end of the video!

Now that we've introduced the concept of AI as a tool to empower people, one example we can explore together is this: How can we use AI to understand and improve representation in social media?

Skin tone emojis are one example of representation in social media. Check out this video with your team before you get started on the skin tone emoji AI activity.

Ready to try this on your own? Follow the steps below:

1. Make a Copy of our Google Sheet: Copy Week 5: Emoji Skin Tone Dataset

2. Watch the video and try the activity with your own copy of the Google Sheet, and answer the reflection questions.

3. When you are done, share your creations with us!

2. • Click Share and Select “Anybody with the Link” can View

   • Submit to our Week 5 Google Form

How does the "AI" work?

... it works by mixing data points! Kind of like mixing paint colors on a palette.

We call this a Linear Model. Linear models are used regularly in data science to make predictions and identify correlations.

• “AI” = computer models that learn from data

• AI models can be written with or without code

• AI models can suffer from bias, but they can also be used to fight bias

After you watch the Takeaways video, scroll down to see the Data Modeling lab, and then move on to Data Games Prep!

Congratulations!! You've finished Boot Camp.

YOU are amazing. And you are well on your journey to becoming a Young Data Scientist.

Now it's time for your group transition to Build and keep working on your Story Project for the Data Games!

Don't worry - you have time! Your project isn't due until April 2nd, and in the next few weeks we will provide even more guidance and inspiration.

Just remember: Your Project is based on data; it tells a story you want to tell to an audience you choose; and it has a positive impact. But what you actually decide to do or 'build' is up to you. Maybe it's...

• An art project you run with elementary students

• A proposal to a university or school district for a class that should be taught

• A new TED Talk series created and run by youth

• Etc. Be creative! :)

In short, your Story-Project can be anything you want it to be...

...as long as it uses data correctly, is built collaboratively, and it has an impact.

Oh by the way, what might impact look like? Something that...

• Improves your community

• Solves a problem

• Raises awareness of an issue

• Makes something more fun, ETC!