Cluster
Cluster documents by embedding similarity to discover themes
Configuration
Clustering algorithm: 'hdbscan' (auto K, recommended), 'kmeans' (requires n_clusters), 'dbscan', 'agglomerative', 'spectral', 'gaussian_mixture'.
Number of clusters (required for kmeans, spectral, agglomerative).
Minimum docs to form a cluster (HDBSCAN/DBSCAN).
Feature URI to cluster on. Auto-detected from upstream feature_search if not set.
Output format: 'clusters' (K summaries), 'labeled' (N docs with cluster_id), 'representatives' (K best docs).
Max members per cluster in output.
Output
No output yet
Configure the stage parameters and click "Run Stage" to see the simulated output.
Related to cluster
Pages on the same stages, extractors and patterns.
Aggregate
Compute aggregations (COUNT, SUM, AVG, etc.) on pipeline results
Group By
Group documents by field value for decompose/recompose workflows
Sample
Sample a subset of documents using random or stratified sampling
Temporal
Group documents by time windows and compute per-window aggregations with drift detection
Limit
Truncate results to a maximum count with optional offset for pagination
Deduplicate
Remove duplicate documents by field match or content similarity