Name: News Article Clusterer — TF-IDF + Hierarchical Clustering Live
Author: MySimulator

Corpus

Documents: 30 Vocabulary overlap / noise: 0.30

Clustering

Linkage method Dendrogram cut height: —

Live stats

Documents—

Vocabulary size—

Clusters at cut—

Avg intra-cluster cohesion—

Points are documents projected onto the top 3 principal components of their real TF-IDF vectors (PCA via power iteration). Drag the 3D view to rotate. The lower panel is the actual dendrogram built by agglomerative merges — drag the cut-height slider to recolor clusters live.