I was wondering if it does makes sense or not to use directed graphs for node classification tasks in graph neural network.
Most of the successful methods for node classification tasks are the ones that deals undirected graphs, even if its original graphs are directed.
Some node classification methods can deals with directed graphs but none of them are more accurate than existing methods which deals undirected graphs.
There are several reasons for this result, but I think the main reason is that the undirected graphs are symmetric and the problem is easy to handle.
So does the edge direction affect accuracy in the node classification task?