Assignment-MATLAB Course: Machine Learning in Finance Question: Research the topics PCA, Kmeans and Heirarchical clusters. Summarize briefly in your own words 2 applications of Kmeans and Heirarchical...

K-Means Cluster
K-means clustering can be defined as a partitioning method. The method partitions the data into k mutually exclusive clusters and the index of the cluster is returned to which the observation is assigned. Each observation is treated as an object in the data which has location in the space. The objective of function is to find a partition in which the objects of that cluster are close to each other as much as possible and as far as possible from other objects of another cluster (Venkateswaran, N., and YV Ramana Rao). It also helps to find out the number of clusters before clustering them.
It is the simplest algorithm for clustering and hence most used. It is an iterative algorithm which partition the dataset into k non-overlapping subgroups.
Applications
1) Image Compression
K-means can be used to compress an Image. We can reduce the number of colors used in an image. Every pixel is used as a data point. Centroid is marked for every cluster.
2) Geyser’s eruptions segmentation
K-means can be used on 2D dataset. The data contains waiting time...