{"id":8015,"date":"2020-12-29T00:26:25","date_gmt":"2020-12-29T00:26:25","guid":{"rendered":"https:\/\/healinglifespan.com\/data-science\/2020\/12\/29\/how-k-means-clustering-algorithm-works\/"},"modified":"2020-12-29T00:26:25","modified_gmt":"2020-12-29T00:26:25","slug":"how-k-means-clustering-algorithm-works","status":"publish","type":"post","link":"https:\/\/wealthrevelation.com\/data-science\/2020\/12\/29\/how-k-means-clustering-algorithm-works\/","title":{"rendered":"How K-Means Clustering Algorithm Works"},"content":{"rendered":"
\n
\"K-means\"K-means<\/span><\/div>\n
\n

\u00a0 In today\u2019s world, where machine learning models implementation is so easy to find anywhere over the internet. It becomes paramount for all machine learning enthusiasts to get their hands dirty on topics related to it.\u00a0<\/p>\n

There are many fascinating topics of supervised and unsupervised learning<\/strong><\/a> or even reinforcement learning come up. But my favorite is the k-means clustering algorithm.\u00a0<\/p>\n

As the name suggests, it is a clustering algorithm<\/strong>.<\/p>\n<\/div>\n

\n
\n
\n

Learn the popular clustering algorithm k-means clustering along with the implementation in python. #datascience #unsupervisedlearning #machinelearning #kmeansclustering #python<\/p>\n<\/p><\/div>\n

\n\t\t\t
\n\t\t\t\t<\/i>
\n\t\t\t\tClick to Tweet<\/span>
\n\t\t\t<\/span>\n\t\t<\/p>\n<\/p><\/div>\n<\/div>\n

\n

Even if you don\u2019t know what is clustering, still, it is ok.<\/p>\n

By the end of this article, you will learn everything you need to know about k-means clustering<\/strong>.<\/p>\n

After reading this article, you don\u2019t need to brush up on k-means clustering topics before attending any data scientist job interview<\/strong><\/a>.<\/p>\n

Excited to learn \ud83d\ude42<\/p>\n

We are too \ud83d\ude42<\/p>\n

Great, before starting the article, let\u2019s look at the topics you are going to learn in this article. Only if you read the complete article \ud83d\ude00<\/p>\n

I am not kidding. It\u2019s true. It will give you a better idea about the entire article flow.\u00a0<\/p>\n<\/div>\n

\n

Let\u2019s split the k-means clustering into two parts,\u00a0<\/p>\n

Let\u2019s learn clustering first<\/strong><\/a>. Then we will use this knowledge to understand k-means clustering.<\/p>\n

What is clustering?<\/h2>\n<\/div>\n
\"What\"What<\/span><\/div>\n
\n

A cluster is a group of similar entities that are kept together<\/strong>. Their similarity decided by the feature they possess and how closely associated compared with the other entities to this feature.\u00a0<\/p>\n

Let\u2019s say we have two points<\/strong> in a 2-d \u00a0graph. Using the euclidean distance, we can measure how close these two points are located.\u00a0<\/p>\n

Likewise, using various similarity measures<\/strong><\/a>, we can find how close\/similar the data points are.\u00a0<\/p>\n

All similar data points form clusters or groups. Creating these clusters in a meaningful<\/strong> way is called clustering.<\/p>\n

In the machine learning world<\/strong><\/a>, clustering is the process in which we segregate a heap of data points into clusters on the basis of their features.\u00a0<\/p>\n

We will, discusses these features in the upcoming sections.\u00a0<\/p>\n

Clustering Real Life Example<\/h3>\n<\/div>\n
\"Clustering\"Clustering<\/span><\/div>\n
\n

One good real-life example for clustering is the world map<\/strong>. If you see here, each color represents a cluster. These clusters are created based on meaningful similarities.\u00a0<\/p>\n

For now, let\u2019s say this similarity is distance<\/strong>. If you take any place in the cluster, it is closer to the center of that cluster compared with other clusters.\u00a0<\/p>\n

This is one of the main rules for creating clusters using any clustering algorithms.<\/p>\n

Any point in the cluster should be closer to that cluster\u2019s center and far from any other cluster.<\/p>\n

In a more technical way, we can say the intra distance<\/strong> between the same points should be smaller compared with the inter points<\/strong> distance of different clusters.<\/p>\n