id	author	title	date	pages	extension	mime	words	sentences	flesch	summary	cache	txt
cord-024499-14jlk5tv	Balalau, Oana	SubRank: Subgraph Embeddings via a Subgraph Proximity Measure	2020-04-17		.txt	text/plain	3741	265	63	We show that our subgraph embeddings are comprehensive and achieve competitive performance on three important data mining tasks: community detection, link prediction, and cascade growth prediction. A cascade graph is sampled for a set of random walks, which are given as input to a gated neural network to predict the future size of the cascade. In [3] , the authors propose an inductive framework for computing graph embeddings, based on training an attention network to predict a graph proximity measure, such as graph edit distance. Given a graph G = (V, E) and set of subgraphs of G, S = {S 1 , S 2 , · · · , S k }, we learn their representations as dense vectors, i.e. as embeddings.	./cache/cord-024499-14jlk5tv.txt	./txt/cord-024499-14jlk5tv.txt