rnew icon6Grab Deal : Flat 20% off on live classes + 2 free self-paced courses! - SCHEDULE CALL rnew icon7

What is K-Means Clustering in Data Science?

Unsupervised learning algorithms can be used to "learn" patterns hidden inside unlabeled data sets to elucidate similarities or regularities that may be present. Clustering and association are two examples of common tasks that may be performed unsupervised. Clustering methods, such as K-means, organize items in a dataset such that they are more comparable to other objects in the same cluster as opposed to objects in a different cluster to facilitate the process of discovering commonalities within the dataset. Using criteria such as minimum distances, density of data points, graphs, or other statistical distributions, the data is clustered into clusters to make it easier to analyze. Understanding k-means clustering in data mining begins with understanding data science; you can get an insight into the same through our Data Science training.   

What is K-means Clustering?

Clustering is the process of grouping similar themes together to locate them in a body of material, whether it is a document or search results. K-means helps organize data into clusters by grouping data points that are similar and lowering the mean distance between geometric points. To do this, it splits the datasets into a predetermined number (the K) of non-overlapping subgroups (or clusters) and assigns each data point to the cluster with the mean cluster center closest to it. This process is carried out in an iterative manner.

When looking for unlabeled data sets, it is usual practice to use K-means as a clustering algorithm because of its widespread use. It is now used in a broad variety of commercial situations, including but not limited to putting together collections of photographs or color palettes that are similar to one another. You can also check out the data science certification guide to understand more about the skills and expertise that can help you boost your career in data science and data transformation in data mining. 

The process of identifying data that varies from the typical pattern is called anomaly detection. In semi-supervised learning, clusters are used with a reduced labeled data set, and supervised machine learning to give more accurate predictions. This type of learning can be more effective than traditional supervised learning.

How Does K-means Clustering Work?

K-means clustering is a popular unsupervised machine learning algorithm that groups similar data points together. It is commonly used in various fields, such as data analysis, image segmentation, and market research. In this guide, we will discuss the step-by-step process of working with K-means clustering.

Step 1: Choose the Number of Clusters

The first step in K-means clustering is to decide on the number of clusters you want to create. This decision can be based on prior knowledge or techniques like the elbow method or silhouette score.

Step 2: Initialize Centroids

Next, randomly select k data points from your dataset as initial centroids for each cluster.

Step 3: Assign Data Points to Their Nearest Centroid

For each point in your dataset, calculate its distance from all k centroids and assign it to the nearest centroid's cluster.

Step 4: Recalculate Centroids

After assigning all data points to their respective clusters, recalculate the centroid for each cluster by taking the mean value of all its assigned points' coordinates.

Step 5: Repeat Steps 3-4 Until Convergence

Repeat steps three and four until there are no more assignment changes or when a maximum iteration limit is reached.


K-means. The k-means algorithm for partitioning, where each cluster’s center is represented by the mean value of the objects in the cluster. 


  • K: The number of clusters, 
  • D: A data set containing n objects.


  • A set of K clusters. 


(1) Arbitrarily choose k objects from D as the initial cluster centers; 

(2) Repeat

(3) (Re)assign each object to the cluster to which the object is the most similar, based on the mean value of the objects in the cluster;

(4) Update the cluster means, i.e., calculate the mean value of the objects for each cluster; 

(5) Until no change;

How Do I Decide What "K Number of Clusters" to Use?

K-means clustering relies on forming extremely efficient clusters to get the desired results. However, determining how many clusters to use is a difficult process. This article discusses the best approach to determining the value of K or the ideal number of clusters. You can find the procedure outlined below:

  • Using Your Elbow

Finding the best cluster size using the Elbow approach is a common practice. The approach in question relies on the WCSS value idea. The sum of all the differences in a cluster is called the WCSS, or the Within Cluster Sum of Squares. Below is the algorithm for determining the worth of WCSS (for 3 clusters):

The WCSS formula shown above includes the

The first term, denoted by Pi in Cluster1 distance(Pi C1)2, is the square sum of the distances between each data point and the centroid inside cluster1, and the second and third terms are the same.

Any distance measure, such as the Euclidean or Manhattan, may be used to determine how far apart data points are from the centroid.

Elbow Technique

from sklearn.cluster import KMeans

cs = []
for i in range(1, 11):
kmeans = KMeans(n_clusters = i, init = 'k-means++', max_iter = 300, n_init = 10, random_state = 0)
plt.plot(range(1, 11), cs)
plt.title('The Elbow Method')
plt.xlabel('Number of clusters')


from sklearn.cluster import KMeans

kmeans = KMeans(n_clusters=2,random_state=0)
labels = kmeans.labels_
# check how many of the samples were correctly labeled
correct_labels = sum(y == labels)
print("Result: %d out of %d samples were correctly labeled." % (correct_labels, y.size))
print('Accuracy score: {0:0.2f}'. format(correct_labels/float(y.size)))

When given N samples X, the k-means method clusters them into K distinct groups C, each of which is defined by its own mean j. Commonly, the means are referred to as cluster centroids.

The K-means algorithm uses the inertia, or within-cluster sum of squared criteria, to choose which centroids to select.

The following are the steps taken by the elbow technique to determine the best possible number of clusters:

  • K-means clustering is carried out on a provided dataset with varying K values (ranges from 1-10).
  • Determines the WCSS value for each possible value of K.
  • The WCSS values computed are plotted as a function of the cluster size (K).
  • When the plot takes on the shape of an arm, with a sharp bend at its end, this is taken as the optimal value for K.

K-means Clustering in Python        

# This Python 3 environment comes with many helpful analytics libraries installed

# It is defined by the kaggle/python docker image: https://github.com/kaggle/docker-python

# For example, here's several helpful packages to load in 

import numpy as np # linear algebra
import pandas as pd # data processing, CSV file I/O (e.g. pd.read_csv)
import matplotlib.pyplot as plt # for data visualization
import seaborn as sns # for statistical data visualization
%matplotlib inline

# Input data files are available in the "../input/" directory.

# For example, running this (by clicking run or pressing Shift+Enter) will list all files under the input directory

import os
for dirname, _, filenames in os.walk('/kaggle/input'):
for filename in filenames:
print(os.path.join(dirname, filename))

# Any results you write to the current directory are saved as output.

from sklearn.cluster import KMeans
kmeans = KMeans(n_clusters=2, random_state=0) 

For data clustering, the KMeans method divides samples into n equal-variance groups by minimizing the inertia or within-cluster sum-of-squares. Inertia, also known as the sum of squares criteria within a cluster, is a common way to evaluate a cluster's internal cohesion.

cta10 icon

Data Science Training

  • Personalized Free Consultation
  • Access to Our Learning Management System
  • Access to Our Course Curriculum
  • Be a Part of Our Free Demo Class


K Means clustering is a powerful unsupervised learning algorithm that can help businesses and organizations make sense of large datasets. By grouping similar data points together, it can reveal patterns and insights that might have been otherwise hidden. However, it's important to remember that k Means clustering is just one tool in the data scientist's toolbox and should be used alongside other techniques for maximum effectiveness.

Ultimately, whether you're working with customer segmentation data or trying to identify anomalies in your network traffic logs - k Means clustering can unlock new insights and drive better decision-making across a wide range of industries. So why not give it a try today? Who knows what secrets might be hiding within your own dataset waiting to be discovered? Learn about neural network guides and python for data science if you are interested in further career prospects in data science. 

Trending Courses

Cyber Security icon

Cyber Security

  • Introduction to cybersecurity
  • Cryptography and Secure Communication 
  • Cloud Computing Architectural Framework
  • Security Architectures and Models
Cyber Security icon1

Upcoming Class

-1 day 23 Feb 2024

QA icon


  • Introduction and Software Testing
  • Software Test Life Cycle
  • Automation Testing and API Testing
  • Selenium framework development using Testing
QA icon1

Upcoming Class

6 days 01 Mar 2024

Salesforce icon


  • Salesforce Configuration Introduction
  • Security & Automation Process
  • Sales & Service Cloud
  • Apex Programming, SOQL & SOSL
Salesforce icon1

Upcoming Class

0 day 24 Feb 2024

Business Analyst icon

Business Analyst

  • BA & Stakeholders Overview
  • BPMN, Requirement Elicitation
  • BA Tools & Design Documents
  • Enterprise Analysis, Agile & Scrum
Business Analyst icon1

Upcoming Class

0 day 24 Feb 2024

MS SQL Server icon

MS SQL Server

  • Introduction & Database Query
  • Programming, Indexes & System Functions
  • SSIS Package Development Procedures
  • SSRS Report Design
MS SQL Server icon1

Upcoming Class

-1 day 23 Feb 2024

Data Science icon

Data Science

  • Data Science Introduction
  • Hadoop and Spark Overview
  • Python & Intro to R Programming
  • Machine Learning
Data Science icon1

Upcoming Class

0 day 24 Feb 2024

DevOps icon


  • Intro to DevOps
  • GIT and Maven
  • Jenkins & Ansible
  • Docker and Cloud Computing
DevOps icon1

Upcoming Class

3 days 27 Feb 2024

Hadoop icon


  • Architecture, HDFS & MapReduce
  • Unix Shell & Apache Pig Installation
  • HIVE Installation & User-Defined Functions
  • SQOOP & Hbase Installation
Hadoop icon1

Upcoming Class

6 days 01 Mar 2024

Python icon


  • Features of Python
  • Python Editors and IDEs
  • Data types and Variables
  • Python File Operation
Python icon1

Upcoming Class

7 days 02 Mar 2024

Artificial Intelligence icon

Artificial Intelligence

  • Components of AI
  • Categories of Machine Learning
  • Recurrent Neural Networks
  • Recurrent Neural Networks
Artificial Intelligence icon1

Upcoming Class

0 day 24 Feb 2024

Machine Learning icon

Machine Learning

  • Introduction to Machine Learning & Python
  • Machine Learning: Supervised Learning
  • Machine Learning: Unsupervised Learning
Machine Learning icon1

Upcoming Class

13 days 08 Mar 2024

 Tableau icon


  • Introduction to Tableau Desktop
  • Data Transformation Methods
  • Configuring tableau server
  • Integration with R & Hadoop
 Tableau icon1

Upcoming Class

6 days 01 Mar 2024