AI BootCamp (Module 1 Crash Course)
Machine learning model that can sort images
Natural Language Processing 7
Machine learning model that can recognise natural language commands
Recommendation Systems 6
Machine learning model that can recommend the reading age of a book based on data about the book
Decisions and Ethics 4
Presentation or report summarising key points
Machine Learning Algorithms 12
In this session we will be looking at some of the algorithms that make machine learning possible.
(Optional) Python and Orange 3
Data visualiastions using Orange Python code for importing data and running machine learning algorithms (decision trees and kNN)
Building a recommendation system
In this project you will make a school librarian character that can make reading book recommendations.
When you describe a book to your machine learning model (e.g. how many pages/pictures it has), it will try to predict who that book might be suitable for.
You will teach the computer to recognise fiction books of different reading levels by giving it examples of each.
Activity – collecting training data
Find examples of fiction books of different reading levels, and collect the following information about them:
- Number of pages in the book
- Number of lines in each page (choose a typical full page of text)
- Number of pictures in the book (if the book is too long, or there are too many pictures to count, it’s okay to make an estimate)
- The reading level (e.g. Key Stage/Accelerated Reading level)
Do this for at least five books at each reading level.
Update the minimum and maximum for each of the sliders so that the range is more sensible.
For example, you could use minimum of 0 and maximum of 300 for pages.
You could use 0 – 40 for number of lines.
You could use 0 – 50 for pictures.
But choose what you think makes sense based on your books.
Choose different numbers – Instead of number of pages, number of lines, and number of pictures, what other numbers could you use?
- The height of the book?
- The thickness?
- The size of the letters?
Try creating a new numbers project and this time use your own ideas. Compare it with your first project – is it better or worse at making recommendations?
How else could you use this technology? Make notes in your workbook.
This Digital Technologies Institute experiment shows how a neural network is trained to produce recommendations based on the data about the books. Work through the steps outlined on the webpage to explore how this works.
- Decision making and recommendations
Write down Possible ideas & Training data required
This contact form is available only for logged in users.