Dunn's method

Dunn's method is a statistical approach used in machine learning to evaluate how well a clustering algorithm separates different groups within data. It calculates a score based on two factors: the smallest distance between points in different clusters (inter-cluster separation) and the largest distance between points within the same cluster (intra-cluster cohesion). A higher Dunn index indicates better-defined, more distinct clusters. Essentially, it helps assess the quality of clustering results by measuring how compact each group is and how far apart the groups are from each other.