LambdaLabTM
Artificial Intelligence · Class 10 · Unit 1 · AI Project Cycle
class 10 aiproject cycle⏱️ 5 min read

Stage 4 — Modelling

Up to here you have been preparing. This is the stage where the project gets the piece that actually decides something — the piece people mean when they say “the AI”.

1What a model is

Modelling
In modelling you choose and build the model — the part of the project that takes data in and gives a decision out. You research several models, test them, and the most efficient one becomes the base of your project.
The six stages, as a cycle

Tap a stage to stand on it, or press play and watch the work go round. Drag sideways to turn.

4. Modelling — Choose and build the model — the part of the project that makes the decision.

It answers: Which model fits this problem best?

Choosing a recipe

You have your ingredients and you have looked at them carefully. Now you try two or three recipes with them and keep the one that tastes best. Modelling is trying the recipes; the winner becomes the dish you serve.

2How modelling actually goes

  1. Look at the pattern you found while exploring the data.
  2. Research the models that suit that kind of pattern, and pick a few candidates.
  3. Train each one on your data — show it the examples so it can learn from them.
  4. Compare them and keep the most efficient one.
  5. Build the rest of the project around that model.
What “training” means
Training is showing the model your data over and over until it adjusts itself to fit the pattern. Nobody writes the rule by hand — the model works it out from the examples. That is the difference between a learning-based model and an ordinary program.
Unit 2 is all about this stage
Which model to build — rule based or learning based; supervised, unsupervised or reinforcement; classification or regression — is the whole of Unit 2. Here you only need to know what the modelling stage is for.

3Practice

1
What happens in the modelling stage of the AI project cycle? (Practice)
2 marks
▶Show the answer

In the modelling stage we decide the kind of model to build from the patterns found while exploring the data, test the models that might suit, and select the most efficient one. That model then becomes the base of the AI project and the algorithm is developed around it.

4Check yourself

Quick Check

Which stage decides what type of model the project will use?

Quick Check

After testing several models, which one becomes the base of the project?

← Previous
📊 Stage 3 — Data Exploration
Next →
Stage 5 — Evaluation ✅