The title of Markovian-type model may be applied to any model that strongly relies on the theoretical foundations drawn by the mathematician Andrey Markov (1856-1922), who describes a system with a set of states and transitional probabilities. The idea behind it is as straightforward as it is aged: Markovian models are sustained by Bayes' theorem.

You may ask—why trust such a model rather than younger ones such as neural networks? Even though neural nets are very powerful indeed, they may be too general given some tasks. Moreover, combining models usually enhances the final result. That said, consider adjusting a Markovian model only for the sake of combining it with other models you may already have.

This section will briefly introduce the fundamentals of Markovian models and HMMs, but not before listing the real-world application of Markovian models.