Assignment
In this assignment, consider the following dataset that is reduced from the iris dataset. Use Manhattan distance for this assignment. You don't need the class column for clustering since it is often not known. You need the class column for just plotting the data.
A1 A2 Class
4.5 1.3 2
5.1 1.8 3
1.4 0.2 1
4.5 1.5 2
5.2 2.0 3
1.5 0.2 1
4.7 1.6 2
5.2 2.3 3
1.7 0.4 1
4.6 1.5 2
Part A- K-means Clustering
A1. Plot the data where attribute 1 (A1) and attribute 2 (A2) are two axes. You can use a cross sign (x) for class 1 points, a circle sign (o) for class 2 points, and a star sign (*) for class 3 points. A2. Cluster the data into two clusters A3. Cluster the data into three clusters.
Part B- Agglomerative Clustering
B1. Cluster the data as we did in the class. B2. Assume that we need to split data into two clusters. Then which points would be in cluster 1 and cluster 2. B3. Repeat B2 with three clusters. B4. Compare your results of B2 and A2 and then compare results of B3 and A3. Are they similar?