RWBP: Detects spatial outliers using a Random Walk on Bipartite Graph
a Bipartite graph and is constructed based on the spatial and/or non-spatial attributes of the spatial objects in the dataset. Secondly, RW techniques are utilized on the graphs to compute the outlierness for each point (the differences between spatial objects and their spatial neighbours). The top k objects with higher outlierness are recognized as outliers.
||RANN, igraph, lsa, SnowballC
||Sigal Shaked & Ben Nasi
||Sigal Shaked <shaksi at post.bgu.ac.il>
||GPL-2 | GPL-3 [expanded from: GPL (≥ 2)]