Scribd
Upload
11 views5 pages

Clustering

Uploaded by

baswalganesh1
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
11 views5 pages

Clustering

Uploaded by

baswalganesh1
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 5

Clustering

Clustering

Clustering helps to create a homogeneous group of customers/entities for better


management of customers/entities. Clustering is a divide-and-conquer strategy
which divides the dataset into homogenous groups which can be further used to
prescribe the right strategy for different groups. In clustering,
the objective is to ensure that the variation within a cluster is minimized while
the variation between
clusters is maximized.

Clustering algorithms are unsupervised learning algorithms (classes are not


known a priori).
(Classification problems are supervised learning algorithms, where classes are
known a priori in the training data. Another important difference between
clustering and classification is that
clustering is descriptive analytics whereas classification is usually a predictive
analytics algorithm.
The main objective of clustering is to create heterogeneous subsets (clusters)
from the original dataset such that records within a cluster are homogeneous
and identify the characteristics that differentiate the subsets. For example, if
a company wants to increase its brand awareness to appeal to all its existing
or possible future customers, it must design a campaign. The company can
design a single campaign to address all its customers. But what if its
customers have different characteristics such as varied income, age,
preferences, profession, gender? The same campaign may not appeal to all
of them. The company can think of running multiple campaigns targeting
different customer segments.
How does clustering work?

Clustering algorithms use different distance or similarity or dissimilarity


measures to derive different clusters. The type of distance/similarity
measure used plays a crucial role in the final cluster formation. Larger
distance would imply that observations are far away from one another,
whereas higher similarity would indicate that the observations are similar
Some common distance metrics:
• Euclidean distance
• Minkowski distance
• Jaccard similarity coefficient
• Cosine similarity
• Gower’s similarity coefficient
K means clustering

Hierarchical clustering

You might also like

pFad - Phonifier reborn

Pfad - The Proxy pFad of © 2024 Garber Painting. All rights reserved.

Note: This service is not intended for secure transactions such as banking, social media, email, or purchasing. Use at your own risk. We assume no liability whatsoever for broken pages.


Alternative Proxies:

Alternative Proxy

pFad Proxy

pFad v3 Proxy

pFad v4 Proxy