Mixture Models in R

Mixture modeling can be used when there is an interest in the heterogeneity of populations. Mixture models use familiar probability distributions (e.g. The mixture models employ familiar probability distributions such as Poisson, Gaussian and Binomial to provide a formal, but useful statistical framework for classifications and clustering. The probability of belonging to a specific cluster can be calculated and inferences made about sub-populations. This is an exception from traditional clustering methods. Marketing, for instance, might allow you to group different customer segments to determine their likelihood to buy specific products. You can target them with targeted promotions more effectively. Natural language processing can be used for grouping documents into topics. This will enable you to determine the importance of each topic in all documents. This course will explain Mixture Models and how to use them.