Machine learning constitutes a very active research area across multiple disciplines as the amount of data that can be collected becomes larger and the ability to uncover hidden patterns through algorithms and models easier. However, many machine learning techniques assume that the data are independent of one another, an assumption that is severely violated when data are collected over time and space. The main objective of this project is to develop probabilistic machine learning models to identify patterns in systems with temporal and spatial dependency to then apply them to animal movement and behavior systems.