Modeling adolescents’ perception of cycling safety: A new approach using graph neural networks and street view imagery

Keyphrases

• Network View 100%
• Graph Neural Network 100%
• Adolescent Perceptions 100%
• Street View Images 100%
• Cycling Safety 100%
• Safety Perception 40%
• Built Environment 20%
• Urban Design 20%
• Road Network 20%
• Visual Representation 20%
• Traditional Models 20%
• Cyclists 20%
• Tree Cover 20%
• Perception Survey 20%
• Overall Accuracy 20%
• Belgium 20%
• Spatial Representation 20%
• Spatial Dependence 20%
• Street-level 20%
• Spatial Attributes 20%
• Interpretive Approach 20%
• Spatial Context 20%
• Urban Street 20%
• Ghent 20%
• Environmental Dimension 20%
• Geographic Context 20%
• Visual Perspective 20%
• Color Texture 20%
• Scalable Approaches 20%
• Class Imbalance 20%
• Monotony 20%
• Actionable Insights 20%
• Visual Indicator 20%
• Sustainable Transport Planning 20%
• Road Infrastructure 20%
• Beyond Linear 20%
• Bicycle Use 20%
• Land Use Indices 20%
• Multidimensional Framework 20%
• Explainable Artificial Intelligence 20%
• Linear Assumption 20%
• Street Environment 20%
• Adaptive Loss Function 20%
• Improved Graph Convolutional Network 20%
• XAI Models 20%
• Texture Complexity 20%

Social Sciences

• Neural Network 100%
• Youth 100%
• Explainable Artificial Intelligence 66%
• Cyclists 33%
• Road Network 33%
• Land Use 33%
• Belgium 33%
• Urban Design 33%
• Sustainable Transport Planning 33%
• XAI 33%
• Loss Function 33%
• Graph Convolutional Network 33%

Computer Science

• Graph Neural Network 100%
• Explainable Artificial Intelligence 100%
• Visual Representation 50%
• Spatial Representation 50%
• Class Imbalance 50%
• Traditional Model 50%
• Graph Convolutional Network 50%
• Visual Indicator 50%