Cerrar
What is clustering?
Partitional clustering
Hierarchical clustering
Clustering algorithms
(3 used in this course)
Clustering distinctions
1. Exclusive versus non-exclusive
2. Fuzzy versus non-fuzzy
3. Partial versus complete
4. Heterogeneous versus homogeneous
Centroid
K - means complexity
Well-separated cluster
Center-based cluster
Contiguity- based cluster
Density-based cluster
Shared property clusters
(conceptual clusters)
Pre-processing and post-processing
(clustering)
Strengths of hierarchical clustering
Two main types of hierarchical clustering
Cluster similarity - MIN or Single Link
Cluster similarity - MAX or Complete Link
Cluster similarity - Group average
Cluster similarity - Ward's method
DBSCAN - density
DBSCAN - MinPts
DBSCAN - Border
DBSCAN - Noise
DBSCAN - pros and cons
