Struct MinHashLsh 

pub struct MinHashLsh { /* private fields */ }

MinHash + LSH index for near-duplicate detection.

Supports insert to add documents and query to find candidates that are likely near-duplicates of a given text.

Implementations

impl MinHashLsh

pub fn new() -> Self

Create a new empty MinHash LSH index.

pub fn insert(&mut self, doc_id: u64, text: &str)

Insert a document into the index.

doc_id is a unique identifier for the document. text is the document content to be shingled and hashed.

pub fn query(&self, text: &str) -> Vec<u64>

Query the index for candidate near-duplicates of text.

Returns a list of document IDs that share at least one LSH band bucket with the query. These are candidates — false positives are possible but false negatives are unlikely for high-similarity pairs.

pub fn compute_signature(text: &str) -> MinHashSignature

Compute the MinHash signature for a text string.

pub fn get_signature(&self, doc_id: u64) -> Option<&MinHashSignature>

Get the stored signature for a document, if it exists.

pub fn len(&self) -> usize

Return the number of documents in the index.

pub fn is_empty(&self) -> bool

Return true if the index is empty.

Trait Implementations

impl Default for MinHashLsh

fn default() -> Self

Returns the “default value” for a type.