Struct SearchIndex 

pub struct SearchIndex { /* private fields */ }

In-memory inverted index mapping tokens to weighted (node_id, weight) pairs.

Built from a KnowledgeGraph via SearchIndex::build. Each node contributes tokens from its label, id, and source_file, each with a different base weight plus a degree-based boost.

Implementations

impl SearchIndex

pub fn build(graph: &KnowledgeGraph) -> Self

Build the inverted index from a knowledge graph.

Token weights:

• Label token: 2.0 + ln_1p(degree) * 0.1
• Id token: 1.0 + ln_1p(degree) * 0.1
• Source file token: 0.5 + ln_1p(degree) * 0.1

Note: the same token from multiple fields (label + id) stacks additively, rewarding nodes whose token appears in multiple fields.

pub fn search(&self, terms: &[String]) -> Vec<(f64, String)>

Search for nodes matching any of the given terms.

Each term is tokenized and matched against the index using exact and prefix matching. Scores are aggregated per node. Results are returned sorted by descending score.

Prefix matches receive half the weight of an exact match.