Clustering_24march_2011

Clustering_24march_2 - Clustering TWSChow Jan2010 We must acknowledge Carnegie Mellon University as some of these slides are modified from the CMU

Info iconThis preview shows pages 1–14. Sign up to view the full content.

View Full Document Right Arrow Icon
Clustering  T W S Chow  Jan 2010 We must acknowledge Carnegie Mellon University as some of these slides are modified from the CMU AI Clustering ppt
Background image of page 1

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
2 Outline What is Clustering? Distance Measure General Applications of Clustering Major Clustering Approaches
Background image of page 2
3 What is Clustering? Organizing data into clusters such that there is high intra-cluster similarity low inter-cluster similarity Clustering = unsupervised classification (no predefined classes) Informally, finding natural groupings among objects.
Background image of page 3

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
4 What is Clustering? What is a natural grouping among these objects? Clustering is subjective Simpson's Family School Employees Females Males
Background image of page 4
5 What is Clustering? Typical usage As a stand-alone tool to get insight into data distribution As a preprocessing step for other algorithms
Background image of page 5

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
6 Distance Measure What is Similarity? not easy to answer, but important. The quality or state of being similar; likeness; resemblance; as, a similarity of features. Similarity is hard to define, but… “ We know it when we see it question. We will take a more pragmatic approach . The real meaning of similarity is a philosophical
Background image of page 6
7 Cosine distance Cosine distance is a widely used measure to reflect the Euclidean distance between two investigated objects. Turn to cosine distance pdf
Background image of page 7

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
8
Background image of page 8
9 Their similarities in this case can be measured by means of dress, height, hair, age, smoke etc. This is a topic of feature extraction.
Background image of page 9

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
10 Dendogram (for hierarchical) clustering A dendogram is a cluster tree diagram In dendogram it shows where split or merge occurs It is a visualization of hierarchical clustering It enables us to specify the cutting pt for determining the number of clusters i.e., Fig. 1, we cut at 2 and obtain 2 clusters {4 objects (3,5,6 4), and 2 objects (1,2)} In Fig. 2, set to 1.2, we obtain 3 clusters. Fig. 1 Fig. 2
Background image of page 10
11 Another case clustering the blue from the reds ( hierarchical clustering )
Background image of page 11

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
12
Background image of page 12
13 General Applications of Clustering Pattern Recognition and Image Processing Data Analysis create a nice visualizing map by clustering feature spaces
Background image of page 13

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
Image of page 14
This is the end of the preview. Sign up to access the rest of the document.

This note was uploaded on 04/14/2011 for the course EE 4146 taught by Professor Tommychow during the Spring '11 term at City University of Hong Kong.

Page1 / 38

Clustering_24march_2 - Clustering TWSChow Jan2010 We must acknowledge Carnegie Mellon University as some of these slides are modified from the CMU

This preview shows document pages 1 - 14. Sign up to view the full document.

View Full Document Right Arrow Icon
Ask a homework question - tutors are online