machine learning guides

What is Machine Learning – An Introduction

Welcome to this introduction to machine learning. Machine learning is a tool used to achieve artificial intelligence. This is done using statistical methods to let computers find patterns in large amounts of data. We say that the machine “learns” rather than being programmed. The terms, machine learning and artificial intelligence, are often used interchangeably. This article outlines the main differences between ML and AI.

Learning happens by training a model. To “train”, you need data. Typically, the data set is divided into a training set and a test set. You train the model on the training set, and then you test whether the model has learned something by testing it on the test set. The test data is data the model has not seen before, and it will reveal whether it has “learned” what it is supposed to. If the model has only memorized training data, it will not do well when tested.

Applications of Machine Learning

Machine learning is now used in everything from self-driving cars, improved web search, email filtering, image recognition, language recognition, enhanced compilers and an expanded understanding of the human genome. Click here to read more about when to use – and when not to use – machine learning.

The two common types of Machine Learning

Machine learning is often divided into two catagories: Supervised learning, non-supervised learning.

Supervised learning

The machine learn a model from historical examples. It learns to understand that certain input values result in certain output values.

A simple example is to classify two or more things using image recognition. The example of cats and dogs is almost always used here, so let’s do birds and dogs for a change. Dogs have four legs, while birds just have two. In addition, the birds have wings. When the model sees one animal with four legs, it will therefore assume it is looking at a dog. Likewise, if the animal has two legs and two wings, it is most likely a bird. This is called classification.

This type of machine learning models have had a lot of success in recent year, and there are even cases where the models outperform humans.

Non-supervised learning

In the case of non-supervised learning, the machine does not have access to output values for given input values. Instead, the algorithm itself attempts to find the structure of the input values, for example, by grouping them into clusters. Us humans are extremely good at this. In many cases, machine learning still has a long way to go to achieve super-human performance.


I hope you liked our short introduction to machine learning and that you are ready to learn even more. This site contains a lot of guides and information about machine learning and artificial intelligence. A good place to start would be these machine learning beginner’s guides.