id	author	title	date	pages	extension	mime	words	sentences	flesch	summary	cache	txt
cord-024504-p2vxnn9z	Lyu, Tianshu	Node Conductance: A Scalable Node Centrality Measure on Big Networks	2020-04-17		.txt	text/plain	4036	277	63	Moreover, with the help of node embedding techniques, Node Conductance is able to be approximately calculated on big networks effectively and efficiently. Thorough experiments present the differences between existing centralities and Node Conductance, its outstanding ability of detecting influential nodes on both static and dynamic network, and its superior efficiency compared with other global centralities. Random walk, which Node Conductance is based on, is also an effective sampling strategy to capture node neighborhood in the recent network embedding studies [10, 21] . We visualize the special case, football network, in order to have an intuitive sense of the properties of Degree, Betweenness, and Node Conductance (other centralities are presented in the Supplementary Material). Comparing with the existing global centralities, Node Conductance computed by DeepWalk is much more scalable and capable to be performed on big datasets. We also rethink the widely used network representation model, DeepWalk, and calculate Node Conductance approximately by the dot product of the input and output vectors.	./cache/cord-024504-p2vxnn9z.txt	./txt/cord-024504-p2vxnn9z.txt