Levels

Levels

• Graph-level

  • Classification

    • Categorize different graph

    • e.g. Molecule property prediction

  • Generation

    • e.g. Drug discovery

  • Evolution

    • e.g. gPhysical simulation

  • Examples

    • Drug discovery

      • Nodes: Atoms

      • Edges: Chemical bonds

    • Antibiotic Discovery

    • GCPN

    • Physics simulation

      • Nodes: Particles

      • Edges: Interaction between particles

• Subgraph-level (community)

  • Clustering

    • Detect if nodes from a community

    • e.g. social circle detection

  • Examples

    • Traffic prediction

      • Nodes: Road segments

      • Edges: Connectivity between road segments

      • Prediction: Time of arrival (ETA)

• Node-level

  • Classification

    • Predict a property of a node

    • e.g. Categorize online users / items

  • Examples

    • AlphaFold

      • Key idea: spatial graph

      • Nodes: Amino acids in a protein sequence

      • Edges: Proximity between amino acids (residues)

• Edge-level

  • Link prediction

    • Predict whether there are missing links between two nodes

    • e.g. Knowledge graph completion

  • Examples

    • PinSage

      • Recommender systems

      • Nodes: users and items

      • Edges: user-item interactions

    • De novo

      • Drug side effects

      • Nodes: Drugs and proteins

      • Edges: Interatctions