Module Check
You made it to the end of Predicting What Comes Next — amazing work! Across this module, you went from learning what a prediction even is, all the way to understanding how machines use millions of examples and confidence scores to predict things every second of every day. Before you go, let us look back at all the big ideas you have collected. Ready? Let us go!
Flashcards — click each card to reveal the answer
Module Quiz — Show What You Know!
What is a prediction?
You notice that your cat always meows right before it rains. Rain is coming — and the cat starts meowing. What are you using to make your prediction?
A machine is taught to recognize apples by studying only 3 photos. What will probably happen when it sees a new apple?
A prediction based on great clues turns out to be wrong. What does this mean?
A machine says it is 97% confident a photo shows a strawberry. What can you conclude?
Which of these is a real-world example of a prediction being made by a machine?
Look at everything you know now: predictions use clues. Patterns help us predict. Machines learn from examples. More varied examples mean better predictions. Predictions can be wrong and that is okay. Confidence tells us how sure a prediction is. Predictions are hiding inside weather apps, games, keyboards, and more. You now understand one of the most powerful ideas in all of computer science. That is something to be proud of!
My Prediction Journal
- Take out a piece of paper and make your very own Prediction Journal for the week.
- Each day for five days, write down two predictions you made that day — about anything: weather, what someone will say, what will be for dinner.
- Write the clues you used for each prediction.
- At the end of the day, write whether each prediction was right or wrong.
- At the end of the week, count your score. How many did you get right?
- Reflect: did you get better at predicting as the week went on? What helped you most — better clues, noticing patterns, or having more past experiences to draw from?