Familiarise yourself with the Teachable machine application. View the supporting videos.
Note: Teachable machine requires Google Chrome on Windows or Macintosh (tablets are not supported) and a webcam.
Image 1: Teachable Machine Application Happy/Sad project screenshot
The image above shows the view of a project created in the Teachable machine AI application. On the left, the classes are shown Background, Happy and Sad. The preview on the left shows the model being tested using a smiley emoji. The AI recognises the emoji as ‘Happy’ (100%) shown as a complete red bar.
Privacy and personal information: Discuss the potential misuse of personal images when uploading images of ourselves or friends on websites. Instruct students not to record images of themselves or others via the webcam or uploading images.
The conditions of use for Teachable machine state that images are not stored on external servers if the teachable machine program is closed when completed and the project is not saved. If students close the tab, nothing is saved in their browser or on any servers.
By the end of this lesson students will:
Use a suitable hook to discuss emotions. Why is it important that we recognise someone’s emotions?
Write or draw some emotions on cards, one emotion per card. Give a student a card to act out. Then ask the class to guess the emotion. Alternatively, show character images from a picture book, and have students guess the emotions.
Ask the class: What tells us that a person is happy, sad, angry or surprised? What is a neutral expression?
As a class, create a table of information to look for patterns in the data.
Example table (yours may be different)
Compare your descriptions of features to emojis or emoticons. How do your descriptions compare? What features are used to convey each emotion?
This activity can be done as a teacher-led activity, or adapted as a hands-on activity to suit a range of skills. Students can begin at Level 1 and demonstrate understanding at each level or start at a selected level.
The level of difficulty could be increased as follows and presented using a gaming analogy. When explaining the task use a combination of verbal and visuals.
The activity has been levelled to enable differentiation.
You may wish to have students do a sorting activity where they have different emoji pictures that they need to cut out and sort into categories.
A sample ‘Sort the emojis’ handout in Word format can be downloaded here.
Have each student create an ‘emoji quiz’ by having them draw out several emojis on a piece of paper and include a word bank. Students then switch papers and try to guess the emotion for each emoji. Students switch back and grade the paper.
A sample ‘emoji quiz’ handout in Word format can be downloaded here.
Have students identify differences between a pair of emojis expressing the same general emotion. For example, show two different versions of a sad face emoji and have students find the differences in the two images (e.g. ‘one has a tear next to its eye’). Also have students decide which of the two images is ‘more’ of that emotion (e.g. ‘more sad’).
A sample ‘More or less emojis’ handout in Word format can be downloaded here.
Model how to create, train and test an AI model that can recognise happy and sad. Refer to this video explanation or follow these steps:
Image 2: Teachable Machine Application screenshot
Image 3: Teachable Machine Application Happy/Sad project screenshot
Talk about the bar on the interface, which displays the AI’s confidence level.
For younger students, discuss the scale. If the bar is fully coloured, the AI is very sure. If the bar is only partly coloured, the AI is not quite sure. Here’s an example from our Happy/Sad AI model:
|Correct: 100% sure||Correct: quite sure||Incorrect|
|The model is working as expected. The emoji takes up the whole screen as it was trained.||The view includes the background and a smaller view of the emoji.||The model is not working as expected. The emoji is small and does not take up the whole screen. The model is showing a bias based on size as it was trained on full screen images. Retraining would include images of different sized emojis.|
AI is the ability of machines to mimic human capabilities in a way that we would consider 'smart'.
Machine learning is an application of AI. With machine learning, we give the machine lots of examples of data, demonstrating what we would like it to do so that it can figure out how to achieve a goal on its own. The machine learns and adapts its strategy to achieve the goal.
In our example, we are feeding the machine images of emojis via the inbuilt camera. The more varied the data we provide, the more likely the AI will correctly classify the input as the appropriate emotion. In machine learning, the system will give a confidence value; in this case, a percentage and the bar filled or partially filled, represented by colour. The confidence value provides us with an indication of how sure the AI is of its classification.
This lesson focuses on the concept of classification. Classification is a learning technique used to group data based on attributes or features.