Learn why and where K-Means is a powerful tool
Clustering is a very important part of machine learning. Especially unsupervised machine learning is a rising topic in the whole field of artificial intelligence. If we want to learn about cluster analysis, there is no better method to start with, than the k-means algorithm.
Get a good intuition of the algorithm
The K-Means algorithm is explained in detail. We will first cover the principle mechanics without any mathematical formulas, just by visually observing data points and clustering behavior. After that, the mathematical background of the method is explained in detail.
Learn how to implement the algorithm in Python
First we will learn how to implement K-Means from scratch. That means for the beginning no additional packages will be used, except numpy. This is important to get a really good grip on the functioning of the algorithm.
You will of course also learn how to implement the algorithm really quickly by using only one line of code.
The examples will be based on artificial data, which we generate ourselves in the course.
Learn where you should pay attention
K-Means is a powerful tool but it definetely has drawbacks! You will learn where you have to be careful and when you should use the algorithm, and also when it is a bad idea to use the algorithm. I will show you examples and counterexamples on the quality and applicability of this method.