search

apple2373 / metapath2vec

185 stars 59 forks source link

metapath strategy: strictly or non strictly? #5

Open chenhui-bupt opened 6 years ago

chenhui-bupt commented 6 years ago

for v ∈ V do MP = MetaPathRandomWalk(G, P, v, l) ; X = HeterogeneousSkipGram(X, k, MP) ; end if the type of v is not equal to the first of your input metapath P, then you will skip it and use the next node v or choose the neighborhood node of v in its appropriate type in P. For example, the metapath is "author-conf-author" and node v is a conf, i wonder should i choose one author node of conf v according the partial path "conf-author" of the metapath P? or we skip the node v until we get an author node to match the metapath "author-conf-author" strictly ? I appreciate your answer on it, Thanks!

chenhui-bupt commented 6 years ago

if the random walk must starts with the node belong to the first type in metapath P? can you tell why? I hope I can express my problem clearly?

apple2373 commented 6 years ago

You should ask the author of the original paper. I think it's reasonable that the random walk must start with the first type of the meta-path, but I don't know the answer for sure. My implementation does NOT include the random walk generator. The input to my implementation assume you already have random walks following the meta-path.