Hello, and welcome to Lesson One.

This lesson is going to introduce the general concept of clustering,

which is an unsupervised machine learning technique,

along with discussed some examples of where it can be applied in business.

So, by the end of this lesson you should understand the basic concepts behind clustering.

That is what it is we're trying to do,

when we apply a clustering algorithm to data.

You should be able to articulate the importance of clustering for

business applications and should be able to give a few examples of those,

and you should be more familiar with some of

the advanced applications of clustering such as customer segmentation.

There are several readings for

this particular lesson and I'm going to walk you through a few of these,

so that you get a feel for what it is I want you to get out of these readings.

The first reading is about customer segmentation,

the idea is when you have a business that has lots of customers.

You may be able to categorize those customers into different groups based on

their behavior or buying patterns and if you understand how each group reacts,

you might be able to more effectively target them in a certain manner.

So for instance, if you can find a group of hype buying customers you might want to

give them incentive to come to your store more often so that they spend

even more and those for instance that may not spend a lot,

you may try to understand why they're not buying

your product and figure out ways to market to

them in order to actually encourage them to shop your store and to make purchases.

So, this is actually a really interesting approach to

the application of cluster finding to a business and I think it's useful to see

this as an application for clustering and understand better why you might

want to spend time exploring a data set in order to figure out,

how can I find customers that are

clustered and then treat all of those customers as one entity.

Another article here talks about how businesses can use

clustering and there's a lot of different ways you can do it,

and so there's nice articles in that it gives some examples of ways that you can use it,

in ways you can think about maybe using clustering in more general sense.

The last two readings are about finding clustering or finding clusters and data.

It's a challenging machine learning topic and this article walks through that.

It's challenging because clustering is generally done in an unsupervised manner,

that is you're given a set of data

with instances and they have values for different features and their data set,

and your goal in cluster finding is to try to use the information and the features to

find data points that are nearby each other in this potentially high dimensional space.

And so, there's algorithms that are designed to do that and they typically

are done in an unsupervised manner and that we don't provide guidance by saying,

we think this is a cluster and this is a cluster.

Instead, we use the data to determine that sense of

similarity or nearby-ness that might be resulting from a distance metric,

being applied to these data for instance.

So this article, first one here talks about why you might want to do clustering,

it talks about some of the different techniques

for clustering such as hierarchical clustering.

There's also a spatial or density based, spatial clustering,

there's also a density based clustering and so these different ways build on each

other and allow you to see

relationships between data that you might otherwise not have seen.

The last article is just the second part of

that first article and this talks a little bit more about,

what's the characteristic of a good cluster and

why you might want to try to find them and

understanding different techniques for

finding the clusters and what their benefits and disadvantages are.

So, with that I'll go ahead and stop this particular video,

I want to encourage you to ask questions in the form if you have them.

Cluster finding is a very interesting and important machine learning topic,

so I encourage you to dig into this deeply during

this particular module and if you have questions let us know and of course, good luck.