Applied Network Science

Table 3 RMSE scores (best scoring in bold) for the Graph Neural Network and Benchmark baseline on forecasting strain sensor readings in the correlation & DTW graph

From: A computational framework for modeling complex sensor network data using graph signal processing and graph neural networks in structural health monitoring

Algorithm	Sensor type
	Non-filtered			Filtered
	X-strain	Y-strain	Combined	X-strain	Y-strain	Combined
Correlation
Benchmark	0.49	0.79	0.58	0.36	0.59	0.45
T-GCN	0.38	0.61	0.44	0.29	0.46	0.36
DTW
Benchmark	0.49	0.79	0.58	0.34	0.49	0.37
T-GCN	0.38	0.61	0.44	0.28	0.39	0.31

Back to article page