ATPGNN: Reconstruction of Neighborhood in Graph Neural Networks With Attention-Based Topological Patterns.
Kehao WangHantao QianXuming ZengMozi ChenKezhong LiuKai ZhengPan ZhouDapeng WuPublished in: IEEE Access (2021)
Keyphrases
- neural network
- neighborhood graph
- connected components
- graph structure
- random walk
- pattern recognition
- graph structures
- graph data
- graph representation
- topological features
- minimal spanning tree
- high resolution
- training patterns
- image reconstruction
- three dimensional
- affinity graph
- structural patterns
- spanning tree
- graph mining
- bipartite graph
- weighted graph
- graph theory
- feed forward
- neural network model
- pattern mining
- fuzzy logic
- training process
- graph matching
- dynamic graph
- back propagation
- data mining techniques
- artificial neural networks
- neighborhood structure
- smooth manifold
- topological map
- graph patterns
- topological properties
- reconstruction method
- directed acyclic graph
- self organizing maps
- genetic algorithm