UN-CCDs extend Cluster Catch Digraphs by using nearest-neighbor-distance Monte Carlo tests instead of Ripley's K to determine covering radii, yielding competitive performance on moderate-dimensional data with complex clusters and uniform noise.
Classification of imbalanced data with a geometric digraph family.The Journal of Machine Learning Research, 17(1):6504– 6543, 2016
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2025 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Clustering with Uniformity- and Neighbor-Based Random Geometric Graphs
UN-CCDs extend Cluster Catch Digraphs by using nearest-neighbor-distance Monte Carlo tests instead of Ripley's K to determine covering radii, yielding competitive performance on moderate-dimensional data with complex clusters and uniform noise.