CLUSTER ANALYSIS

Prepared By: (Group-5)

Ashish Goyal
Jitendra Jain
Nitesh Sadani
 Definition
• “Cluster analysis helps us to identify the
groups or segments that are more likely to
each other then they are like members of
other groups or segments”

• “It is the technique that is used to classify

objects In to relatively in to homogenous
groups called clusters.”
 The diagram below illustrates the results of a survey that studied
drinkers’ perceptions of spirits (alcohol). Each point represents the
results from one respondent. The research indicates there are four
clusters in this market.
Purpose for cluster analysis
• Segmenting the market
• Understanding buyers behavior
• Identify new market opportunities
• Selecting test markets
• Reducing data
Statistical terms
• Agglomeration Schedule
• Cluster Centroids
• Cluster Centers
• Cluster membership
• Dendrograms
• Distance between cluster centers
• Icicle diagram
• Coefficient matrix
Procedure
Formulate the problem
●

Select a distance measure

●

Select a clustering procedure

●

Decide on the number of clusters

●

Interpret the profile clusters

●

Assess the validity of clustering

●
Formulate the problem
• Select the variables that you wish to apply
the clustering technique to.
• The set of variables should show some
similarity or difference between
themselves.
• Variables should be selected on the basis
of past research theory.
• The sample size should be taken as large
as possible.
Select a distance measure

• The selected data should have either some

similarity or difference.

–There are various ways of computing distances:

–Euclidean distance - the square root of the sum of
the squared differences in value for each variable
–Manhattan distance - the sum of the absolute
differences in value for any variable
–Chebyshev distance - the maximum absolute
difference in values for any variable
 Select a clustering procedure

Clustering Procedures

Hierarchical Non - Hierarchical Others

Agglomerative

Sequential Parallel Optimizing

Li nkage Method Variance Method
Divise
CentroidMethods

Threshold Threshold Partioning Two step

Sources : Marketing Research/Naresh Malhotra/Cluster Analysis/P-643
Sources : Marketing Research/Naresh Malhotra/Cluster Analysis/P-644
Decide on number of clusters
• Theoretical, practical or conceptual consideration
may suggest a certain no. of clusters.
• The distance at which clusters are combined can
be used as the criteria. (Hierarchical)
• The elbow shaped curve indicates the no. of
clusters. (Non - hierarchical)
• The relative size of the clusters should be
meaningful.
Interpret and profile the clusters
• Use of illustrative techniques like:
• Dendrograms
• Icicle plots
• Perceptual maps
Assess Reliability and Validity
• repeat analysis but use different distance
measure
• repeat analysis but use different clustering
technique
• split the data randomly into two halves and
analyze each part separately
• repeat analysis several times, deleting one
variable each time
• repeat analysis several times, using a
different order each time
 Example
• We consider a clustering of consumers based on
attitudes towards shopping. Consumers were asked for
their degree of agreement on a 7-point scale (1=disagree,
7=agree):

• V1 : shopping is fun
• V2 : shopping is bad for your budget
• V3 : I combine shopping with eating out
• V4 : I try to get the best ways when shopping
• V5 : I don’t care about shopping
• V6 : You can save a lot of money by comparing
prices
Sources : Marketing Research/Naresh Malhotra/Cluster Analysis/P-641
 Cluster Centroids
Means of Variables
Cluster V1 V2 V3 V4 V5 V6
No.

1 5.750 3.625 6.000 3.125 1.750 3.875

2 1.667 3.000 1.833 3.5000 5.5000 3.333

3 3.500 5.833 3.333 6.000 3.500 6.000

Sources : Marketing Research/Naresh Malhotra/Cluster Analysis/P-648

Other clustering applications
• Social network analysis
• Image segmentation
• Data mining
• IMRT segmentation
• Grouping of Shopping Items
• Petroleum Geology
• Crime Analysis
 References
• Marketing Research/Naresh Malhotra/Ch-
20/P-637
• http://www.wikipedia.com\Cluster analysis (in
marketing) - Wikipedia, the free
encyclopedia.htm
• http://www.scribd.com/cluster/2341235.htm
 Feel Free to revert me at
ashish.goyal3@gmail.com

Thank You