Playground survey - Exploring machine learning

micro:bit - the next gen > playground survey

You can no longer submit your playground survey data, but you can still do all of the activities with your class.

An inspiring, active and engaging whole class data collection activity that uses the micro:bit machine learning tool to demonstrate how real-world data samples are used to trainmachine learning models. Pupils answer the questions:

How can computers learn from data? How can machine learning models be trained ‘better’?

Watch the video

• To complete this survey activity you will need the helpful teacher instructions.
• This activity is primarily a Computing/ICT lesson that will build on the themes and ideas from the Tracking our activity sessions and dive a little deeper into machine learning and how it works. Pupils will be introduced to an online machine learning tool created by the Micro:bit Education Foundation.
• They will learn/be reminded that the micro:bit has a built-in sensor called an accelerometer that can be used to measure when the micro:bit moves in different directions. They will also learn how to create their own machine learning model that will be able to recognise the difference between specific movements.
• This activity will help pupils understand some of the underlying principles of AI and give them the knowledge they need to be well informed and able to make good decisions about using new technology. It offers some positive ideas for the inclusive use of AI technology and could get pupils thinking about how AI could help solve real world problems.
• We would love to see how your class is getting involved with the BBC micro:bit playground survey. Why not share updates about your activities on social media and let us know by tagging @BBC_Teach and using #BBCplaygroundsurvey

• Science: Using equipment to measure and record data; Collecting data and using it to answer a scientific question.
• Maths and Numeracy: Reading and interpreting data in the form of table, charts and graphs.
• Design and Technology: Testing a product's ability to fulfil a design brief and/or solve a problem.
• PHSE/RSE/Health and Wellbeing: Data privacy and security; Evaluating digital information – accuracy, honesty and credibility.
• Computing/ICT: Using various forms of input and output including sensors; Recording, storing, organising and analysing data; Using technology responsibly and safely - see PSHE/RSE/Health and Wellbeing.

• Introduction and planning: To understand that sensors can transmit data in real time to a program stored on a computing device (Computing/ICT).
• Exploring and training: To learn how data can be used to create a model that can then be used to represent the real world; To understand that an AI machine learning system uses a model to help it make decisions about new data (Computing/ICT).
• Discussion: To use what they learn about data collection and processing to make better decisions about technology; To develop confidence in reading and interpreting data in the form of graphs and tables (Computing/ICT/Maths and Numeracy).

• Science: Find out more about the physical structure of the hand and arm by looking at X-rays and diagrams; Explore the mechanics of waving and clapping by observing muscle and joint activity; Use the micro:bit accelerometer and machine learning tool to explore other aspects of the human body eg. leg movement when running, walking and jumping.
• PHSE/RSE/Health and Wellbeing: Revise understanding of personal and private data and how to keep it secure; Discuss issues around consent when providing data to online tools, platforms and companies – read through sample data consent notices (eg. a cookie notice) and create their own clearer versions; Learn about how keeping personal data anonymous can support data privacy and security whilst still allowing the data to be used.
• Design and Technology: Design a gadget or device that uses the machine learning model you have created; Consider how AI and machine learning models could be added to existing devices and systems – what sort of training data would it need – eg. how to make the perfect cup of hot chocolate.
• Computing/ICT: Discuss how machine learning and AI are changing the way we use technology – using familiar examples; Explore the issues around ‘gaps’ in training data and how this might make a model that was biased or inaccurate – look at some real life scenarios and examples.

More activities from the playground survey