Graph Algorithms¶

Uni includes a built-in graph algorithm library exposed through Cypher procedures. Use it to compute centrality, clustering, paths, and connectivity without exporting data.

What It Provides¶

• Centrality: PageRank, betweenness, closeness, eigenvector, Katz.
• Connectivity: WCC, SCC, bridges, articulation points.
• Paths: shortest paths, k-shortest, A*, BFS/DFS.
• Community detection: Louvain, label propagation.

Example¶

use uni_db::Uni;

# async fn demo() -> Result<(), uni_db::UniError> {
let db = Uni::open("./my_db").build().await?;
let session = db.session();

let rows = session.query(
    "CALL uni.algo.pageRank() YIELD nodeId, score RETURN nodeId, score ORDER BY score DESC LIMIT 10"
).await?;

println!("{:?}", rows);
# Ok(())
# }

import uni_db

db = uni_db.Uni.open("./my_db")
session = db.session()
rows = session.query(
    "CALL uni.algo.pageRank() YIELD nodeId, score RETURN nodeId, score ORDER BY score DESC LIMIT 10"
)
print(rows)

Use Cases¶

• Identify important nodes in a graph.
• Detect communities and clusters.
• Compute routing or similarity metrics.

Use from Locy FEATURES¶

Ten of these algorithms are also callable as Locy feature expressions inside CREATE MODEL, so a classifier can consume topology directly without a separate feature pipeline:

CREATE MODEL risk_scorer AS
  INPUT (n)
  FEATURES degree_centrality(n), pagerank_score(n), avg_neighbor(n, 'risk')
  OUTPUT PROB risk
  USING xervo('classify/risk-v1')

See Neural Predicates for the full reference.

When To Use¶

Use built-in algorithms when you need graph analytics without exporting data or building your own pipeline.