Portfolio Assignment 3

Clustering

Summary

Add a .html file to your portfolio titled Cluster Analysis.

Purpose & Requirements

Demonstrate your knowledge of clustering by describing how the three classic algorithms K-Means, Aggl. Hierarchical, and DBSCAN work. In addition, present pieces of your clustering code in action. Your writing should include:

• An introduction that summarizes what we mean by "clustering."
• Brief sections for each clustering algorithm (K-Means, Aggl. Hierarchical, DBSCAN)
• Each section should include the algorithm's title, a brief but accurate description, perhaps an illustration, benefits and drawbacks of each, and common use cases (when might Aggl. Hierarchical be the preferred choice of strategy?)
• Where possible, provide example implementations (your code) of the relevant algorithm.

This document should be equivalent to two or three double-spaced, typewritten pages in length. Be brief, but thorough -- the reader should have a good understanding about the various algorithms, and believe that you have a strong understanding of them.

Grading Criteria (250 points)

Your portfolio must contain a table of contents in index.html and have one link called Cluster Analysis; and one page called clustering.html that fulfills the requirements described above.