Mixture Model

Mixture Model is a type of probabilistic model which are capable of detecting similar groups of data. It is the process of grouping similar data is known as clustering, segmentation or density estimation. Mixture Model use to make statistical inferences about the properties of the sub-populations given only observations on the pooled population, without sub-population identity information. Mixture model can be extended within the Bayesian network framework to create models with additional structure. The additional structure allows it to use a compact probability distribution leading to increased performance, reduced memory consumption and greater interpretability when viewing the model graphically.