介紹

<aside> 💡

本頁面介紹Knowledge Graph，Graph相關資料已移到Graph頁面。

</aside>

• Knowledge
  • The Atoms of Knowledge
• Graph (內容移到Graph)
• GQL (內容移到Graph)
• Knowledge Graphs
  • This is a tutorial based on this excellent and comprehensive tutorial on Knowledge Graphs
• Large Language Models and Knowledge Graphs: Ways to combine them
  • Chain of History (CoH) reasoning
  • Right for Right Response
  • KnowPhish
• Knowledge Graphs & LLMs: Fine-Tuning vs. Retrieval-Augmented Generation
• Impact of Quantization on Knowledge Graphs with Deepseek-R1 and Mistral Models (Friend Link)
• Dependent Types: A New Paradigm for Ontologies and Knowledge Graphs
• TypeDB: A Polymorphic Database for AI Agent Memory and Complex Ontology
  • TypeDB combines concept modeling with dependent type theory to create a mathematically sound foundation for knowledge representation. Dependent type theory extends formal logic with types that can depend on other types, enabling complex relationship modeling.
• Graphiti: Knowledge Graph Memory for a Post-RAG Agentic World

Graph embedding

• What are graph embeddings ?
  • Node embedding techniques
    • DeepWalk
    • Node2Vec
    • GraphSAGE
  • Graph embedding algorithms
    • Graph Convolutional Networks (GCNs)
    • Graph Attention Networks (GATs)
    • Graph Neural Networks (GNNs)
• Knowledge Graph Embeddings: Unraveling the What, Why, and How
  • There are two main approaches to KGE:
    • Distance-Based Embeddings
    • Semantic Matching Embeddings
  • The Landscape of KGEs:
    • TransE
    • RotatE
    • ComplEx
    • TuckER
• Survey on Embedding Models for Knowledge Graph and its Applications
  • Translation-Based Models
    • Translating Embeddings (TransE)
    • Translating Relation (TransR)
    • Distinguished Multilinear (DistMult)
    • Complex Embeddings (ComplEx)
  • Neural Network-Based models
    • Semantic Matching Energy (SME)
    • Multi-Layer Perceptron (MLP)
    • Neural Tensor Network (NTN)
    • Neural Association Model (NAM)
    • Convolutional KB (ConvKB)
    • Relational Graph Convolutional Network (R-GCN)
• A Survey on Knowledge Graph Embedding: Approaches, Applications and Benchmarks
  • Categories of KG Embedding Approaches
    • Triplet Fact-based Representation Learning Models
    • Description-Based Representation Learning Models
  • Translation-Based Models
    • TransE、TransH、TransR、TransD、TransA、KG2E、TransG
  • Tensor Factorization-Based Models
    • RESCAL、DistMult、ComplEx、HolE、SimplE、RotatE、QuatE
  • Neural Network-Based Models
    • SME (Semantic Matching Energy)、NTN (Neural Tensor Network)、MLP (Multi-Layer Perceptron)、NAM (Neural Association Model)、RMNN (Relational Memory Neural Network)、ConvKB (Convolutional Knowledge Base Embeddings)
  • Textual Description-Based Models
    • DKRL (Description-Embodied Knowledge Representation Learning)、TEKE (Type-Embodied Knowledge Embedding)
  • Relation Path-based Models
    • PTransE (Path-based TransE)、RPE (Relation Path Embedding)
• Towards Named Entity Disambiguation with Graph Embeddings (Friend Link)
  • Named Entity Disambiguation (NED) is a critical NLP task that involves resolving ambiguities in entity mentions by linking them to the correct entries in a knowledge base
• Knowledge graph embedding methods for entity alignment: experimental review
  • Methods Used for Entity Alignment
    • Relation-Based Methods
      • MTransE
      • MTransE+RotatE
      • RDGCN (Relational Dual-Graph Convolutional Network)
      • RREA (Relational Reflection Entity Alignment)
• Graph Neural Networks
  • Creating Embeddings from Knowledge Graphs Using Graph Neural Networks