Absolute indices for determining compactness, separability and number of clusters

The paper introduces absolute cluster indices that assess both compactness and separability of clusters, moving beyond relative measures commonly used in clustering validation. It defines a compactness function for each cluster and a set of neighboring points for cluster pairs to evaluate cluster quality and overall distribution margin. These indices are applied to determine the true number of clusters and are compared against widely-used validity indices on synthetic and real-world datasets.

Machine-generated by The Flow from the publisher's headline and feed description — not written or checked by a human. The full article lives at arXiv Machine Learning.