A rating prediction model with cross projection and evolving GCN for bitcoin trading network

The timely rating forecast of customers is crucial for the evaluation of trading service. Due to the high complexity and variability of rating, traditional static graph models have been proved to be ineffective in learning the topological and temporal features of dependent nodes and the representation of edge weights. In this paper, we propose a trading evaluation prediction method on the bitcoin marketplace and a novel deep learning framework to tackle the dependent problem of topological and temporal information. We elaborate a hierarchical framework including dynamic rating construction network layer, extracting node features layer and edge learner layer. Firstly, we obtain the data and organize them into a graph sequence as input. Secondly, we learn the node representation in the extracting node features layer by using the evolving graph neural network to learn the dependency of topological and temporal information. Finally, we use cross projection to learn the edge representation in the edge learner layer. Experimental results show that our model can effectively predict the edge weights and outperforms state-of-art baselines on various real-world transaction datasets, which give a more accurate evaluation in trading evaluation.