Which clustering method uses density-based reachability to construct a reachability-plot for cluster extraction without fixing global $\epsilon$?
OPTICS (Ordering Points To Identify the Clustering Structure) addresses DBSCAN's difficulty with varying density by generating an augmented ordering.
What is the main advantage of BIRCH (Balanced Iterative Reducing and Clustering using Hierarchies) for large datasets?
BIRCH builds a CF Tree summarizing sub-clusters in a single pass, enabling fast hierarchical clustering on massive datasets.
In a Clustering Feature vector CF = (N, LS, SS) for a cluster of N points, what does SS represent?
A CF tuple consists of N (number of points), LS (linear sum sum x_i), and SS (square sum sum $x_i^2)$.
Which distance metric between clusters in CURE (Clustering Using Representatives) makes it robust to non-spherical shapes?
CURE selects well-scattered representative points for each cluster and shrinks them toward the centroid by a factor $\alpha$.
What is the primary difference between Hard C-Means and Fuzzy C-Means (FCM) clustering?
$FCM allows data points to belong to multiple clusters simultaneously with varying membership degrees u_{ij} summing to 1.$