Scribd
Upload
56 views

Soft Clustering

Soft clustering allows data points to belong to multiple clusters, with varying degrees of membership. Two common soft clustering algorithms are fuzzy c-means and rough k-means. Fuzzy c-means assigns each point a probability of belonging to different clusters, while rough k-means defines lower and upper approximations of cluster membership. The rough k-means algorithm aims to minimize uncertainty by representing clusters as rough sets with probabilistic boundaries.

Uploaded by

Arun Vignesh
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
56 views

Soft Clustering

Soft clustering allows data points to belong to multiple clusters, with varying degrees of membership. Two common soft clustering algorithms are fuzzy c-means and rough k-means. Fuzzy c-means assigns each point a probability of belonging to different clusters, while rough k-means defines lower and upper approximations of cluster membership. The rough k-means algorithm aims to minimize uncertainty by representing clusters as rough sets with probabilistic boundaries.

Uploaded by

Arun Vignesh
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
You are on page 1/ 7

Soft Clustering

Introduction to Clustering algorithm

• Clustering is one of the widely used data mining techniques for medical diagnosis.

• Clustering can be considered as the most important unsupervised learning technique

• Basically, Clustering problem is about partitioning a given data set into groups such that points in a cluster are more similar to

each other points in different clusters

• The fundamental objective of clustering is to group similar objects in the same cluster and dissimilar objects in different

clusters

• This is also a method of identifying similar groups of data in datasets

• Types of Clustering

• Hard Clustering

• Classical K-Means or Hard K-Means

• Soft Clustering

• Fuzzy c-means

• Rough K-Means
Fuzzy Clustering
• In this clustering technique, the data points in any of the cluster is not

decisive. Here, one data point can belong to more than one cluster group.

• It provides the outcomes as the probability of the data point belonging to

each of the cluster.

• Algorithm used in Fuzzy clustering is Fuzzy c-means

• It is similar process to the K-means clustering but differs in the

computation like fuzzifier and membership value.


Fuzzy c-means
• This is a method of clustering which allows one piece of data to
belong to two or more clusters.
• Its main objective is to minimize the weight sum of euclidean
distances between the objects i and the means μk of the
corresponding clusters k.

SIDD IROD CIRDD MARD


Ageonset 56.1 34.0 44.7 66.9
BMI 30.1 28.3 40.3 31.9
Waist 103 100 135 107
Hba 10.3 5.9 9.6 9.4
Triglycerides 116 224 694 188
HDL 28 41 20 59
Rough KMeans

• Rough Kmeans cluster is described by two hard


approximations, a lower and upper approximation.

• A cluster is represented by rough set based on a lower


approximation and an upper approximation.

• The rough K-Means algorithm provides a rough set


theoretic flavor to the conventional K-means algorithm
to deal with uncertainty involved in cluster analysis
Rough K-Means Algorithm
• Input – Dataset of (n) objects with (d) feature number of cluster(k),
value of parameter wlower & wupper & Epsilon (Փ)
• Output – Lower approximation u(k) & upper approximation ū(k) of (k)
cluster
• Algorithm
• Step 1: Randomly choose the data object (to at least one lower approximation).
• Step 2: Computing cluster centroids (following equation applied for centroid
clustering).
• Step 3: Assign each object to the lower approximation and upper
approximation to the cluster.
• Step 4: Find the distance from the centroid to each point using Euclidean
distance.
• Step 5: Repeat Step 2 & 3 until convergence (old centroid - new centroid)
Rough K-Means Clustering Output
• Two types of Rough KMeans performed for our 19084 datasets

RoughKMeans_LW performs Lingras & RoughKMeans_PE performs Peters’ k-


West’s k-means clustering algorithm. means clustering algorithm.
Rough KMeans_LW Rough Kmeans_PE
8 Traits 8 Traits
Cluster 1 2 3 4 Cluster 1 2 3 4
Ageonset 47.2 43.0 46.3 44.5 Ageonset 47.2 42.6 46.1 44.1
BMI 27.5 26.8 27.7 27.2 BMI 27.3 26.6 27.7 27.1
Waist 96.1 95.7 96.8 96.3 Waist 95.5 95.5 96.7 96.2
Hba 8.5 9.3 8.8 9.0 Hba 8.4 9.5 8.8 9.0
Triglycerides 114.6 421.3 175.6 264.6 Triglycerides 101.3 485.4 178.8 289.7
HDL 40.7 37.1 39.2 38.0 HDL 41.1 36.9 39.0 37.8
CPF 1.1 1.2 1.2 1.2 CPF 1.1 1.3 1.2 1.2
CPS 2.7 2.6 2.7 2.6 CPS 2.7 2.5 2.7 2.6

Functions Functions
RoughKMeans_LW(dataMatrix, meansMatrix, nClusters, RoughKMeans_PE(dataMatrix, meansMatrix, nClusters,
maxIterations, threshold, weightLower). maxIterations, threshold, weightLower)

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