Reproduce “Persistent Multiscale Density-based Clustering” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Daniël Bot, Leland McInnes, Jan Aerts
PLSCAN tracks HDBSCAN* leaf clusters across minimum-size cutoffs on a hierarchy built with fixed k, then selects the cutoff with the greatest total cluster persistence. At k = 4, its median adjusted Rand index was 0.66, versus 0.57 for HDBSCAN* EOM across labeled datasets. Its median non-noise fraction was lower: 0.76 versus 0.88.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.