Handling Missing Sensors in Topology-Aware IoT Applications with Gated Graph Neural Network.
Shengzhong LiuShuochao YaoYifei HuangDongxin LiuHuajie ShaoYiran ZhaoJinyang LiTianshi WangRuijie WangChaoqi YangTarek F. AbdelzaherPublished in: Proc. ACM Interact. Mob. Wearable Ubiquitous Technol. (2020)
Keyphrases
- neural network
- back propagation
- small world
- real time
- management system
- sensor networks
- random walk
- graph structure
- directed graph
- graph theory
- multilayer perceptron
- graph representation
- structured data
- cloud computing
- fully connected
- artificial neural networks
- topological information
- graph theoretic
- graph model
- network model
- weighted graph
- graph matching
- feed forward
- missing data
- knn
- pattern recognition
- neural nets
- bipartite graph
- recurrent neural networks
- neural network model
- network architecture
- spanning tree
- sensor data
- sensor technology
- prediction model
- missing values
- data fusion