build_dag_taxonomy

build_dag_taxonomy(
    embeddings,
    k_neighbors=30,
    similarity_threshold=0.55,
    diversity_gap_threshold=0.02,
    verbose=False,
)

Build a navigable DAG taxonomy from embedding space.

Creates a lattice structure where: - Nodes with high neighbor diversity are “general” (parents) - Nodes with low neighbor diversity are “specific” (children) - Multiple parents are allowed (non-tree structure)

Parameters

Name	Type	Description	Default
embeddings	np.ndarray	Embeddings to analyze (n, d). Will be L2-normalized.	required
k_neighbors	int	Number of neighbors for k-NN graph.	30
similarity_threshold	float	Minimum similarity for parent-child edges.	0.55
diversity_gap_threshold	float	Minimum diversity difference for directed edge.	0.02
verbose	bool	Print progress information.	False

Returns

Name	Type	Description
	DAGTaxonomy	DAGTaxonomy with navigation methods.

Example

from dyf import build_dag_taxonomy

taxonomy = build_dag_taxonomy(embeddings)

Navigate the taxonomy

ancestors = taxonomy.get_ancestors(query_idx) common = taxonomy.get_common_ancestors(idx_a, idx_b) hubs = taxonomy.get_convergence_points(min_parents=10)