Struct SemanticTermWeighter 

pub struct SemanticTermWeighter { /* private fields */ }

TF-IDF / BM25 term weighter for a corpus of documents.

Maintains document profiles and document-frequency counts so that term weights can be computed incrementally as documents are added or removed.

Implementations

impl SemanticTermWeighter

pub fn new(config: WeighterConfig) -> Self

Create a new weighter with the given configuration.

pub fn add_document(&mut self, doc_id: &str, terms: &[&str])

Register a document by its id and a slice of terms.

Counts each term occurrence and updates corpus-level statistics (document frequencies, average document length).

If a document with the same doc_id already exists it is replaced.

pub fn remove_document(&mut self, doc_id: &str) -> bool

Remove a document from the corpus. Returns true if it existed.

pub fn weight_terms(&self, doc_id: &str) -> Result<Vec<TermWeight>, String>

Compute weights for every term in the specified document.

pub fn tf(&self, term: &str, doc_id: &str) -> Option<f64>

Raw term frequency: count / total_terms.

pub fn idf(&self, term: &str) -> f64

Inverse document frequency: ln((N + 1) / (df + 1)) + 1.

pub fn bm25_score(&self, term: &str, doc_id: &str) -> Option<f64>

BM25 score for a single term in a document.

pub fn similarity(&self, doc_a: &str, doc_b: &str) -> Result<f64, String>

Cosine similarity between the TF-IDF vectors of two documents.

pub fn doc_count(&self) -> usize

Number of documents in the corpus.

pub fn vocab_size(&self) -> usize

Number of unique terms across all documents.

pub fn stats(&self) -> TermWeighterStats

Aggregate statistics snapshot.