Did you know that machine learning and artificial intelligence (AI) are actually two different concepts? It’s very common to confuse the two, so let’s dive deeper into what sets machine learning and AI apart.
What is Machine Learning?
Machine learning is the ability for computers to program themselves to continuously learn from any given dataset (taken from MIT Sloan).
This subfield of artificial intelligence was inspired by how the human brain learns, similar to when you practise solving maths problems using the theory you’ve learned in class to be able to solve new problems during a test. In the same sense, machine learning allows programs to practise predicting all kinds of data in order to accurately predict new data, ranging from customer segmentation to tomorrow’s stock price.
What is Artificial Intelligence?
Artificial intelligence is the result of re-creating the capabilities of the human mind using computers and machines.
As the umbrella term for machine learning, artificial intelligence encompasses the whole capabilities of the human mind, not only the learning aspect. Following the maths test example, artificial intelligence would also involve interpreting the invigilator’s oral instructions as well as the test’s written text. Thus machine learning is only one of the processes involved in artificial intelligence.
Case Study
Let's use a simple case study to learn how to identify machine learning and artificial intelligence based on real-world applications.
A well-known example of artificial intelligence is YouTube's recommendation engine, which is the system that allows YouTube to recommend relatable content to the user. Tying back to machines’ attempts to mimic the human mind, recommendation systems can be thought of as a shopkeeper recommending a new product that a loyal customer might like based on their past purchases, which in the case of YouTube is content.
When you watch videos on YouTube, you’ll notice that new videos continue to show up on your recommended list, making it difficult to stop watching. That is artificial intelligence playing a role in keeping users like you engaged.
YouTube collects your content consumption behaviour data and uses machine learning to understand what kinds of content interest you, and which ones you would most likely watch if recommended. If you click any content on your home feed, YouTube's recommendation engine knows that it's right and will continue providing similar types of content.
This machine learning process is included in the recommendation engine to automatically update so that every time you watch a video, new videos of a similar type will be recommended to you.
Of course this is just a simple explanation for YouTube’s complex artificial intelligence, but if you’re interested in a more detailed explanation, check this article from HackerNoon.
Summary
We've discussed the differences between machine learning and artificial intelligence, which is how one is a subfield of the other, and the applications for both using a YouTube case study. YouTube is just one of the many samples of artificial intelligence available, but there are many more ranging from driverless cars to chatbots.
With that in mind, adirani DS is developing our very own artificial intelligence system to diagnose your business' digital health and we need your help to improve our machine learning models by filling out this quick 5-minute survey. If you’re interested in learning more about this, we’d love to speak with you!