The detection of stellar streams, tidal disruptions and merger remnants within the stellar halo provide observational insight into the formation and evolution history of the Galaxy. Often these stellar over-densities have in the past been detected using visual inspection. However, the expected data volume from Gaia will make detection via visual inspection unfeasible and requires an effective detection methodology. Here we present proof of concept results using a point ordering method to identify clustering (OPTICS). We find that OPTICS is an advantageous clustering algorithm to detect non-convex and irregular shaped streams and clouds while simultaneously detecting small scale-substructure in large scale over-densities.