Hierarchical hidden Markov model

A hierarchical Hidden Markov Model (HHMM) is a way to capture patterns in data that has multiple states. It can be thought of as a kind of map, where each state is a place on the map, and the connections between the states represent how you can get from one place to another. In the HHMM, the connections between the states are weighted so that it looks more like a mountain range than a flat land. The weights on the connections are like hills, so that some connections are easier to travel than others. The HHMM is useful because it can be used to predict patterns in data that have multiple states.