K-means clustering is used in data mining to analyze clusters. Method divides n observations into k clusters, deciding by distance from nearest mean. It tries to find centers of natural clusters in data by iterative improvement. After defining all clusters, clustering
Hidden Markov Models
The easiest way to treat sequential data would be simply to ignore the sequential aspects and treat the observations as independent and identically distributed (i.i.d.)[Bishop 2006] as shown in Figure 1. However, this approach would fail to exploit sequential patterns.