id	author	title	date	pages	extension	mime	words	sentences	flesch	summary	cache	txt
work_zbpzx43flfaifddpftczprcome	Ilya Makarov	Dual network embedding for representing research interests in the link prediction problem on co-authorship networks	2019	20	.pdf	application/pdf	9187	1122	54	network and use its embedding for further generalizing author attributes. graph feature engineering and network embedding methods were combined for co-authorship network embeddings and manually engineered features for HSE researchers. future links based on network topology without any additional information on authors. embeddings for author research interests and node proximity and evaluated different Table 3 Comparing machine learning models based on the Neighbor Weighted-L2 link embedding applied to future links prediction on the Table 5 Comparing machine learning models based on the Neighbor Weighted-L2 link embedding for link prediction problem on the HSE dataset. Table 6 Comparing machine learning models based on the Neighbor Weighted-L2 link embedding for link prediction problem on the Scopus Dual network embedding for representing research interests in the link prediction problem on co-authorship networks Dual network embedding for representing research interests in the link prediction problem on co-authorship networks	./cache/work_zbpzx43flfaifddpftczprcome.pdf	./txt/work_zbpzx43flfaifddpftczprcome.txt