I read that a recommended procedure is to first perform hierarchical clustering and then to use obtained cluster centers as initial cluster centers in k means clustering. I suppose that SPSS does not provide cluster centers from hierarchical clustering. I would appreciate suggestion which program to use, and what is he procedure to obtain cluster centers in hierarchical clustering.
Irena Stojkovic I think python is qualified!
import numpy as np from scipy.cluster.hierarchy import linkage, fcluster from scipy.spatial.distance import pdist data = np.random.rand(10, 2) Z = linkage(data, 'ward') clusters = fcluster(Z, t=3, criterion='maxclust') cluster_centers = [] for i in np.unique(clusters): cluster_data = data[clusters == i] cluster_center = np.mean(cluster_data, axis=0) cluster_centers.append(cluster_center)
You're right! SPSS does not directly provide the cluster centers from hierarchical clustering. However, there are other tools and procedures you can use to first perform hierarchical clustering and then use the obtained cluster centers as initial centers for k-means clustering. Python provides excellent tools for both hierarchical and k-means clustering. You can use R or MATLAB as well.
- Badges
- Science topic
- Similar topics
- Mathematical Sciences
- Control Theory
- Control Systems