A simple iterative grid- and density-based Clustering Algorithm

We introduce Iteridense, an iterative clustering algorithm combining grid-based and density-based methods. It
provides two ways to perform the clustering and for both it provides a clear path on how to change the algorithm’s input
parameters to achieve suitable results. We show that Iteridense provides lower computational complexity than pure
density-based algorithms and that it performs clustering at least as good as the DBSCAN algorithm. Our algorithm is
deterministic, does not forcibly assign all data points to a cluster and optionally allows to specify the number of clusters.
We provide a reference implementation of Iteridense as well as a stand-alone program with a graphical user interface.