Manual

Contents:

• Overview
  • Hypergraphs
  • As language: syntactic rules
  • As knowledge model
  • As database
  • Knowledge Agents
• Semantic Hypergraph notation
  • Hyperedge types
  • Type inference rules
  • Argument roles
  • Subtypes
  • Further type-specific additional information
  • Namespaces
  • Full atom structure
  • Special atoms
• Special relations
  • Coreferences
  • Taxonomies
  • Lemmas
• Basic hypergraph operations
  • The two central functions of Graphbrain: hgraph() and hedge()
  • Creating and manipulating hyperedges
  • Creating and populating hypergraphs
  • Adding many hyperedges as a batch (for speed)
  • The neighborhood of a hyperedge (star)
  • Hyperedges containing a given set of hyperedges
  • Searching for hyperedges
  • Degrees and deep degrees
  • Hyperedge attributes
  • Local and global counters
  • Working with hyperedge sequences
• Patterns
  • Searching with wildcards
  • Non-strict search
  • Matching argroles
  • Patterns with variables for information extraction
  • Functional patterns
• Discovering patterns
  • A simple example
  • Focusing the exploration by controlling expansions
  • Including explicit roots in the patterns
  • Matching subtypes
  • Build your own
• Parsing natural language
  • Lemmas
  • Parser-level coreference resolution and inference of gender, number and animacy
• Readers
• Co-reference resolution
• Meaning extraction
• Hypergraph database backends
  • SQLite 3
  • LevelDB
• Notebooks and visualizations
• Command-line interface
  • Commands
• Internals and extending Graphbrain
  • Creating and training a new parser
  • Creating a new hypergraph database backend
• API reference
  • main functions
  • hyperedge module
  • hypergraph module
  • patterns module
  • processor module
  • memory package
  • parsers package
  • readers package
  • utils package
  • processors package