Electric vehicles are gaining popularity as an environmentally friendly alternative to gasoline-powered vehicles. Predictions of propensity to purchase electric vehicles can provide valuable insights for the automotive industry, policy makers, and energy suppliers. At present, there is no comprehensive quantitative research on the influencing factors of electric vehicle purchase intention, and there is also a lack of explainable correlation between electric vehicle purchase intention and these influencing factors. In this paper, we propose a heterogeneous graph neural network model for electric vehicles purchase propensity prediction. This model uses a heterogeneous graph structure to process various types of data including charging infrastructure, environmental factors, brand recognition, etc., and encodes by analyzing the characteristics of different nodes to achieve different group embedding and obtain the final node embedding. The proposed model outperforms several structures on real-world datasets, demonstrating its effectiveness in electric vehicles purchase propensity prediction.
A Heterogeneous Graph Neural Network Model for Electric Vehicle Purchase Propensity Prediction
2023-10-27
1989531 byte
Conference paper
Electronic Resource
English
Traffic Prediction Using Graph Neural Network
IEEE | 2023
|Effectiveness and Heterogeneous Effects of Purchase Grants for Electric Vehicles
DataCite | 2023
|Electric Vehicle Battery State of Charge Prediction Based on Graph Convolutional Network
Springer Verlag | 2023
|