Discrete Dislocation Dynamics simulations have become an important tool for analyzing the mechanical response of micro-meter sized samples. Scaling these simulations up to larger length and time scales is an complex task where various coarse graining and data mining approaches are helpful tools. In this presentation we show various methods and concepts for extracting information from systems of dislocations that are otherwise not easily accessible. Examples include the investigation of internal energies as input for continuum models on larger length scales and data mining of in-situ TEM data in high-entropy alloys, using both 'classical' statistical methods as well as various machine learning-based approaches.