HITS

SQL function: cugraph_hits

Compute HITS hub and authority scores.

Signature

cugraph_hits(table_name [, src_col, dst_col [, weight_col [, options_json]]])

Allowed argument counts: 1, 3, 4, 5.

Quickstart

SELECT * FROM cugraph_hits('target_edges')

Positional arguments

Argument	Type	Required	Default	Notes
table_name	Utf8	yes
src_col	Utf8	no	src
dst_col	Utf8	no	dst
weight_col	Utf8|null	no		accepted as an edge-column binding; native algorithm execution does not consume weights; semantic effect: none for this algorithm
options_json	Utf8	no

JSON options

Option	Type	Default	Constraints	Description
epsilon	Float64	0.00001	> 0
max_iterations	UInt32	100	min 1
normalize	Boolean	false

Graph construction options

Shared by all cuGraph functions, shown here with this function's defaults. The construction_policy option controls whether Nexus requests Python cuGraph-compatible edge normalization or bypasses it for raw libcugraph-style construction; see graph construction options for the full policy guide.

Option	Type	Default	Constraints	Description
construction_policy	Utf8	"python_cugraph"	one of "python_cugraph", "raw_libcugraph"	Edge-list construction semantics used before calling libcugraph.
directed	Boolean	true		Whether graph construction treats edges as directed.
renumber	Boolean	true		Whether graph construction may renumber external vertex identifiers internally.

Output schema

Column	Type	Nullable	Description
vertex	Int64	no	Vertex receiving HITS scores.
hub_score	Float64	no	HITS hub score for the vertex.
authority_score	Float64	no	HITS authority score for the vertex.

note

These are the generic registry schemas. Run cugraph_validate_call for the concrete, table-specific output schema of a particular call.

Examples

This example runs on the citation network demo dataset.

Hubs are surveys, authorities are foundations

HITS returns two scores per vertex in one pass. In a citation graph they separate two kinds of importance that PageRank blends: a hub cites many authorities (a good survey), an authority is cited by many hubs (a foundational result). One call, two ORDER BYs:

SELECT p.year, p.n_references, p.title
FROM cugraph_hits('citation_edges', 'src', 'dst') h
JOIN papers p ON p.paper_id = h.vertex
ORDER BY h.hub_score DESC
LIMIT 4;

year	n_references	title
2019	292	Deep Learning for Generic Object Detection: A Survey
2018	276	Deep Learning for Generic Object Detection: A Survey.
2015	299	Recent Advances in Convolutional Neural Networks
2019	211	Object Detection With Deep Learning: A Review

SELECT p.year, p.n_citation, p.title
FROM cugraph_hits('citation_edges', 'src', 'dst') h
JOIN papers p ON p.paper_id = h.vertex
ORDER BY h.authority_score DESC
LIMIT 4;

year	n_citation	title
2004	35,541	Distinctive Image Features from Scale-Invariant Keypoints
2014	18,029	VERY DEEP CONVOLUTIONAL NETWORKS FOR LARGE-SCALE IMAGE RECOGNITION
2012	16,802	ImageNet Classification with Deep Convolutional Neural Networks
2005	19,433	Histograms of oriented gradients for human detection

The top hubs are literally titled "Survey" and "Review"; the top authorities are SIFT, VGG, AlexNet, HOG. The mutually-reinforcing definition lands both lists in the field with the densest hub/authority structure — computer vision — without being told any field labels.

Limitations & notes

• dry-run validates table resolution, column presence, static dtypes, and options only
• dry-run does not scan edge data, construct a graph, or prove source-vertex existence

Validate before running

Always dry-run a call before executing it. Validation checks the function, table, columns, dtypes, and options without touching the GPU:

SELECT * FROM cugraph_validate_call(
  'cugraph_hits',
  'your_edges_table',
  '{"src_col":"src","dst_col":"dst"}'
);

See Discovery & validation for the full cugraph_validate_call contract.