Artificial Intelligence: Machine Learning, Neural Networks and Natural Language Processing

Topic: Y3 Artificial Intelligence: machine learning, neural networks, natural language processing

What artificial intelligence is and how it works, explained step by step: machine learning in everyday life; the principles and types of AI, its history, machine learning, neural networks and deep learning, natural language processing, computer vision, and real-world uses in Ghana.

View resource Free to join

What's inside

You can unlock a phone with your face, ask Google Translate to turn English into French, or have Google Maps steer you around a traffic jam in Accra. A few decades ago, all of these needed a human brain. Today, they are done by Artificial Intelligence (AI). This note explains what AI is, the main techniques behind it, and how it is used around you.

SUMMARIZED

Year 2: Machine learning, the part of AI that learns from data, with everyday examples (see 'Machine learning simulation tools' for full detail).
Year 3: The principles and types of AI, a brief history, machine learning, neural networks and deep learning, natural language processing, computer vision, and real-world applications.

Year 2: Machine learning: where AI starts

In Year 2 you met machine learning: teaching computers to learn and improve from data without being told exactly what to do, the way a student improves with practice. Social media feeds, shopping recommendations, voice assistants, spam filters and weather forecasts all use it. Machine learning is a part of AI, and in Year 3 you see the bigger picture.

Year 3: Artificial intelligence

What AI is

Think about how you learn from experience, understand conversations and recognise faces. Artificial Intelligence is the field of computer science that builds smart machines to do tasks that usually need human intelligence. Three ideas sit at its core:

Simulating human intelligence: AI systems don't just follow fixed instructions; they analyse information, find patterns and make decisions.
Learning and adapting: like you getting better by studying past questions, AI improves by processing new data. Google Maps learns about traffic jams from many users and suggests faster routes.
Performing complex tasks: translating languages, diagnosing diseases and driving cars, all once thought possible only for humans.

Types of AI

Type	What it is	Status
Narrow AI (weak AI)	Built and trained for one specific task; very good at it, but can't work outside it. Examples: Siri, ChatGPT, Gemini, YouTube's video recommendations.	The only kind we have built
General AI (strong AI)	A machine as intelligent as a human, able to learn and solve any problem.	Not achieved yet; seen in science fiction
Super-intelligent AI	Smarter than the best human minds in almost every field.	Purely theoretical

A brief history

The term "Artificial Intelligence" was first used in 1956 at a conference at Dartmouth College in the USA. Early AI used rule-based systems to solve problems and play games like chess. Then came "AI winters", periods when progress slowed and funding fell because early promises were too hard to achieve with the technology of the time. In the last couple of decades AI has boomed again thanks to much more computing power, huge amounts of data, and better algorithms.

Open the full resource free with a TudHub account.

More on Artificial Intelligence: machine learning, neural networks, natural language processing

Other resources you can find on TudHub.