TwiddleLab / Courses / Artificial Intelligence

Learning From Data

Turn data into tables, ask the neighbors, grow decision trees and test fairly.

Intermediate · 11 lessons · Free, no account needed

Best after: How a Neural Network Thinks

Start the course

Lessons

1. Data Is a Table (5 min): A machine learns from a dataset: a table of examples.
2. Ask Your Neighbors: kNN (7 min): k-nearest neighbours labels a new example by a vote of the k closest examples in the dataset.
3. Twenty Questions: Decision Trees (6 min): A decision tree answers with a chain of yes or no questions about the features.
4. Train, Test, Don't Peek (7 min): A model learns from the training set and is judged on a test set it has never seen.
5. Accuracy Can Lie (6 min): When one kind is rare, a model can score high accuracy by always guessing the common kind.
6. Where to Draw the Line: Precision and Recall (7 min): A model gives each example a score. The threshold decides who gets flagged.
7. Scale Your Features (7 min): kNN adds up differences across features, so a feature with big numbers (grams) can drown out one with small numbers (centimetres).
8. Finding Groups: Clustering (7 min): Clustering finds groups in data that has no labels: unsupervised learning.
9. Biased Data, Biased Machine (7 min): Data bias: if a group is rare in the training data, the model learns little about it and fails it more often.
10. How Your Feed Picks: Recommendations (7 min): A recommendation system suggests what you'll like from what people like you liked: collaborative filtering, kNN with viewers as points.
11. Project: Be the Data Scientist (8 min): A pipeline: data, split, model, evaluate, report.

Lessons are interactive and need JavaScript. Turn it on to play them.