id	author	title	date	pages	extension	mime	words	sentences	flesch	summary	cache	txt
work_y4fwwb63b5a77gm3wpynicxd4q	Dell Zhang	Probabilistic Verb Selection for Data-to-Text Generation	2018	18	.pdf	application/pdf	9876	1089	70	Probabilistic Verb Selection for Data-to-Text Generation In this paper, we address the problem of verb selection for data-to-text NLG through a principled model for verb selection based on large-scale realworld news corpora, and demonstrate its advantages similar to ours is a corpus based method for verb selection developed by Smiley et al. Given a new percentage change, their method randomly selects a verb verb selection for data-to-text NLG (see Section 1) verbs w, and P(x|w) is the likelihood, i.e., the probability of seeing the percentage change x given that percentage change x is to compute the posterior probability distribution P(w|x) over all the possible verbs Given a new percentage change x, in order to calculate its probability of being generated from a verb that we are aware of to this specific task of verb selection in the context of data-to-text NLG is the method	./cache/work_y4fwwb63b5a77gm3wpynicxd4q.pdf	./txt/work_y4fwwb63b5a77gm3wpynicxd4q.txt