-
Launch 8
Machine learning model that can sort images
-
What is machine learning?Lecture1.1
-
How do machines learn?Lecture1.2
-
[LAB] Building a model in Scratch3Lecture1.3
-
[LAB] How to use (step by step)Lecture1.4
-
Extension – refining and adapting modelLecture1.5
-
Facial recognitionLecture1.6
-
Launch 5 questionsQuiz1.1
-
What next?Lecture1.7
-
-
Natural Language Processing 7
Machine learning model that can recognise natural language commands
-
Natural Language ProcessingLecture2.1
-
ChatbotsLecture2.2
-
[LAB] Build an anti-bullying AI prototypeLecture2.3
-
Test and improve your modelLecture2.4
-
Chatbot applicationsLecture2.5
-
Natural Language Processing 1 questionQuiz2.1
-
What next?Lecture2.6
-
-
Recommendation Systems 6
Machine learning model that can recommend the reading age of a book based on data about the book
-
Recommendation systemsLecture3.1
-
Using a recommendation systemLecture3.2
-
Building a recommendation systemLecture3.3
-
Filter bubblesLecture3.4
-
Deeper dive (extension)Lecture3.5
-
Recommendation Systems 3 questionsQuiz3.1
-
-
Decisions and Ethics 4
Presentation or report summarising key points
-
Self driving carsLecture4.1
-
Bias in Machine LearningLecture4.2
-
Review of Machine LearningLecture4.3
-
Extension – Dangers of Machine LearningLecture4.4
-
-
Machine Learning Algorithms 12
In this session we will be looking at some of the algorithms that make machine learning possible.
-
IntroductionLecture5.1
-
Supervised and unsupervised learningLecture5.2
-
Linear RegressionLecture5.3
-
[LAB] Linear regressionLecture5.4
-
Naive Bayes AlgorithmLecture5.5
-
[LAB] Naive BayesLecture5.6
-
Decision TreeLecture5.7
-
[LAB] Decision TreeLecture5.8
-
k-Means ClusteringLecture5.9
-
[LAB] k-Means ClusteringLecture5.10
-
Neural networksLecture5.11
-
SummaryLecture5.12
-
-
(Optional) Python and Orange 3
Data visualiastions using Orange Python code for importing data and running machine learning algorithms (decision trees and kNN)
-
Machine learning toolsLecture6.1
-
Hello worldLecture6.2
-
Decision trees – iris datasetLecture6.3
-
Test and improve your model
It’s time to test your model with examples that you haven’t shown the computer before.
- If your original example for training was “What an idiot”
- Try testing the model with “You are an idiot” or “I think you are an idiot”
Try new instructions and note down the result – did the model correctly carry out your instructions? Why? Why not?
Test instruction | Outcome | Correct Y/N?
This contact form is available only for logged in users.
If you’re not happy with how the computer recognises the instructions, add some more examples. Make sure you run ‘Train new model’ with the new examples though!
Leave Scratch open and go back to the Learn & Test page in the Training tool. Type a message into the Test box that has nothing to do with kind or mean messages.
The confidence score should be very low. The computer is telling you that it’s not certain it understands your command, because what you typed doesn’t look like what it learned from your examples.
Ideas and Extensions
Instead of just kind and mean messages, can you add another category?
Is 75% the right threshold to use to decide whether the computer has recognised the command?
Experiment with different values until you have a value that works well for your machine learning model.
If you choose a number that is too high, the computer will say “Sorry I’m not sure what you mean” too often. If you choose a number that is too low, the computer will get too many decisions wrong.