Artificial Intelligence Crash Course Part 2 - What is Unsupervised Learning?

“People worry that computers will get too smart and take over the world, but the real problem is that they're too stupid and they've already taken over the world.” 

― Pedro Domingos

If you have missed the part 1 on Supervised learning, should read that first!

What is Unsupervised Learning?

Unsupervised learning is where you only have input (data) and no corresponding output (variables). The aim for unsupervised learning is to model the underlying structure or distribution in the data in order to learn more about the data.

Why is it called unsupervised learning? Because unlike supervised learning, in unsupervised learning has no correct answers and there is no teacher that supervises! These algorithms on their own discover and present the interesting structure in the data. (eg, explores customer demographic data to identify patterns)

Unsupervised learning problems can be grouped into clustering and association problems.

• Clustering: A clustering problem is where you want to discover the inherent groupings in the data. An example will be …grouping customers by purchasing behaviour. 
• Association:  An association rule learning problem is where you want to discover rules that describe large portions of your data. An example will be… finding out people that buy something also tend to buy another thing. 

When should you use Unsupervised Learning?

You do not know how to classify the data, and you want the algorithm to find patterns and classify the data for you.

How does Unsupervised Learning work? 

The algorithm receives unlabelled data (For example, a set of data describing customer journeys on a website)

It infers a structure from the data: The algorithm identifies groups of data that exhibit similar behaviour (For example, forms clusters of customers that exhibit similar buying behaviours)

There is semi Supervised Learning type too and this is when solving problems where you have a large amount of input (data) and only some of the data is labelled are called semi-supervised learning problems.

These problems sit in between both supervised and unsupervised learning. For our conversation, we do not need to go deeper into this. 

Some Business Use Cases of Unsupervised Learning

• Detecting digits and letters from hand written copy 
• Customer Segmentation 
• Visual Recognition  

(more coming in Crash Course Part 3 soon!)

Above crash course article is taken from the 'Executive Guide to Artificial Intelligence' curated by Tejas Oza. 

Source : DigitalizaitonTech