Intelligent Data Analysis
SUSTech Summer Semester
Course Material and Useful Links
Peter Tino
P.Tino@cs.bham.ac.uk

< Back to homepage

Demonstrations

Videos might need some time to be loaded.

Table of Contents

Online Clustering Cluster Number Selection Self-Organizing Map - "Bicycle Chain" Self-Organizing Map - "Fishing Net" k-Nearest Neighbors

Online Clustering

Online Clustering

Note: the video is sped up every iteration to save time

[MATLAB codes]

Cluster Number Selection

Part 1

Cluster number = 1,2,3,4

[MATLAB codes]

Part 2 (continued)

Cluster number = 5,6,7

Self-Organizing Map - "Bicycle Chain"

Iteration 1

[MATLAB codes]

Iteration 2,3 (continued)
Iteration 4,5,6 (continued)

Self-Organizing Map - "Fishing Net"

By Guoji

Iteration 1

[MATLAB codes]

Iteration 2,3 (continued)
Iteration 4,5,6 (continued)

k-Nearest Neighbors

Considering a binary classification problem. We have 6 datasets in which the 2 classes of data are more and more overlapping from dataset 1 to 6.

For each dataset, we visualize the classification results when k = 1,3,5,7,9.

How does the decision boundary change as k changes?

[MATLAB codes]