The introductions of Graph2Vec(at the last) and GL2Vec are almost same in this page. I think those are about GL2Vec, cuz the main idea of Graph2Vec is similar to doc2vec.
Here's a sentence from the source paper(page 4, Figure 1), may it be useful:
'''
Given a graph G, it samples c rooted subgraphs around different nodes that occur in G and uses them analogous to doc2vec’s context words and thus learns G’s representation.
'''
No. It is correct - see : "First, the algorithm creates the line graph of each graph in the graph dataset." This is the main difference between these two algos. Have a nice day!
The introductions of Graph2Vec(at the last) and GL2Vec are almost same in this page. I think those are about GL2Vec, cuz the main idea of Graph2Vec is similar to doc2vec.
Here's a sentence from the source paper(page 4, Figure 1), may it be useful: ''' Given a graph G, it samples c rooted subgraphs around different nodes that occur in G and uses them analogous to doc2vec’s context words and thus learns G’s representation. '''
Thanks for your projects, helped me a lot!