VOUW is both a theoretic framework and an algorithm that discovers spatial structure in 2D-data by using a novel compression method. Our theory is based on the Minimum Description Length (MDL) principe that in turn is a practical approximation of Kolmogorov-complexity. Both principles are used extensively in the field of data mining and explanatory data analysis.

This repository is home to the ongoing implementation of libvouw as well as QVouw, a graphical interface to libvouw based on Qt5.