graphrag_core/nlp/mod.rs
1//! Advanced NLP Module
2//!
3//! This module provides advanced natural language processing capabilities:
4//! - Semantic chunking algorithms
5//! - Custom NER training pipeline
6//! - Syntax analysis
7//!
8//! ## Features
9//!
10//! ### Semantic Chunking
11//! - Multiple chunking strategies (sentence, paragraph, topic, semantic, hybrid)
12//! - Intelligent boundary detection
13//! - Coherence scoring
14//! - Configurable chunk sizes and overlap
15//!
16//! ### Custom NER
17//! - Pattern-based entity extraction
18//! - Dictionary/gazetteer matching
19//! - Rule-based extraction with priorities
20//! - Training dataset management
21//! - Active learning support
22//!
23//! ### Syntax Analysis
24//! - Part-of-speech tagging
25//! - Dependency parsing
26//! - Noun phrase extraction
27pub mod custom_ner;
28pub mod semantic_chunking;
29pub mod syntax_analyzer;
30
31// Re-export main types
32pub use semantic_chunking::{
33 ChunkingConfig, ChunkingStats, ChunkingStrategy, SemanticChunk, SemanticChunker,
34};
35
36pub use custom_ner::{
37 AnnotatedExample, CustomNER, DatasetStatistics, EntityType, ExtractedEntity, ExtractionRule,
38 RuleType, TrainingDataset,
39};
40
41pub use syntax_analyzer::{
42 Dependency, DependencyRelation, NounPhrase, POSTag, SyntaxAnalyzer, SyntaxAnalyzerConfig, Token,
43};