Ravi Prakash Gupta

PGDB-10-10-22
Institute of Bioinformatics and
Applied biotechnology,
Bangalore.

Fuzzy C-Means 1
Types of Clustering
 Exclusive Clustering (K-Means)
 Overlapping Clustering (Fuzzy C-Means)
 Hierarchical clustering
 Mixture of Gaussians or Probabilistic
Clustering

Fuzzy C-Means 2
“Fuzzy clustering is a process of
assigning these membership levels,
and then using them to assign data
elements to one or more clusters.”

Fuzzy C-Means 3
A fuzzy set is a pair (A,m) where A is a set and .
m:A[0,1]
For each X belongs to A , m(x) is called the grade
of membership of x in (A,m). For a finite set A =
{x1,...,xn}, the fuzzy set (A,m) is often denoted
by {m(x1) / x1,...,m(xn) / xn}.
Let X€A . Then x is called not included in the fuzzy
set (A,m) if m(x) = 0, x is called fully included if
m(x) = 1, and x is called fuzzy member if 0 <
m(x) < 1. The set {x€a/m₍x₎>0}is called the
support of (A,m) and the {x€A|m(x)=1}set is
called its kernel.

Fuzzy C-Means 4
 Fuzzy c-means (FCM) is a method of clustering
which allows one piece of data to belong to two
or more clusters using Euclidean Distance .

 This method was developed by Dunn in 1973

and improved by Bezdek in 1981 and it is
frequently used in pattern recognition.

Fuzzy C-Means 5
 Step-1: Start

 Step-2: Take n number of Similar points.

 Step-3: Assign threshold value for each

cluster.

 Step-4: Take the ‘point coefficient’ or the

degree to which the point is associated with
the cluster.
.

Fuzzy C-Means 6
 Step-5: Check point coefficient for each point by
iterating the magnitude of the distances

 Step-6: Repeat step-5 to compute :

a) Centroid for each cluster
b) Point coefficient of being in the
cluster.

 Step-7: Repeat step-6 until the algorithm

minimizes intra-cluster variance as well.

 Step-8: Stop

Fuzzy C-Means 7
 With fuzzy c-means, the centroid of a cluster is
computed as being the mean of all
points,weighted by their degree of belonging to
the cluster.
 The degree of being in a certain cluster (degree
of membership) related to the inverse of the
distance to the cluster.
 By iteratively updating the cluster centers and
the membership grades for each data point, FCM
iteratively moves the cluster centers to the
"right"location within a data set.

Fuzzy C-Means 8
 Medical diagnostics : Example- MRI.

 Libraries: book ordering.

 Marketing: finding groups of customers with

similar behavior given a large database.

 Insurance: identifying groups of motor insurance

policy holders with a high average claim cost;
identifying frauds.

Fuzzy C-Means 9
CASE STUDY
 In this work, unsupervised clustering methods
were performed to cluster the patients into
three clusters by using thyroid gland data
obtained by Dr.Coomans.

 Here, Fuzzy C-Means (FCM) and Hard Cmeans

(HCM) algorithms are used as an unsupervised
clustering method to cluster the patients.

 As a result of clustering algorithms, patients’

statuses are classified normal, hyperthyroid
function and hypothyroid function
 The application of Fuzzy C-Means causes the class
membership to become a relative one randan object can
belong to several classes at thesame time but with
different degrees. This is an important feature for
medical diagnostic systems to increase the sensitivity.
 The graphical representation of the results is shown
below
 For the membership degrees
close to 0.5 are the
suspicious cases (shaded area
in Table) to assign the
sample to one cluster.
Therefore fuzzy c-means
clustering for medical
diagnostic systems is more
reliable than the hard one.
 In medical diagnostic
systems, fuzzy c-means
algorithm gives the better
results than hard-kmeans
algorithm.
 Fuzzy Means clustering in Medical Diagnostics.

 www.endocrineweb.com/thyroid.html

 www.mathworks.com

 http://home.dei.polimi.it/matteucc/Clustering
/tutorial_html/cmeans.html

 http://home.dei.polimi.it/matteucc/Clustering
/tutorial_html/AppletFCM.html
Fuzzy C-Means 13