Clustering and Classification

Presented by:

Yogendra, Govinda, Lov, Sunena

Outline

• Background

• Classification

• Clustering

• Examples

• References

Background

• Clustering is “the process of organizing objects into groups whose members are similar in some way”.

• A cluster is therefore a collection of objects which are “similar” between them and are “dissimilar” to the objects belonging to other clusters.

Clustering

Clustering

• Organizing objects into group.

Cluster analysis Components

• Similarity(Distance) Measure

• Clustering Algorithm

Similarity Measure

• Distance between two data points

Clustering Algorithms• Exclusive Clustering• Data belongs to only one definite cluster

• Eg. K-Means Algorithms

• Overlapping Clustering• Uses fuzzy set for clustering

• Single Data may belong to one or more clusters

• Hierarchical Clustering• Agglomerative(Bottom-up)

• Divisive(Top-Down)

Clustering Algorithms

Agglomerative

Divisive

K-Means Algorithm

K-Means Algorithm

Minimization of Squared Error Function

Demo

Here we have a dataset!

We randomly choose 2 group centroids!

We assign each point to the group that has the closest centroit.

We recalculate the positions of the centroids.

We assign each point to the group that has the closest centroid.

We recalculate the positions of the centroids.

Other Approaches

Hierarchical clustering

• Agglomerative (bottom up)

1. start with 1 point (singleton)

2. recursively add two or more appropriate clusters

3. Stop when k number of clusters is achieved.

Hierarchical clustering

• Divisive (Top down)

1. Start with a big cluster

2. Recursively divide into smaller clusters

3. Stop when k number of clusters is achieved.

References

• www-users.cs.umn.edu/~kumar/.../chap8_basic_cluster_analysis.pdf

• http://en.wikipedia.org/wiki/Cluster_analysis

• https://www.cs.duke.edu/courses/fall03/cps260/notes/lecture18.pdf

• http://www.matlab-cookbook.com/recipes/0100_Statistics/150_kmeans_clustering.html

• http://www.cs.utah.edu/~germain/PPS/Topics/Matlab/plot.html

• http://www.mathworks.com/help/stats/k-means-clustering.html