id	author	title	date	pages	extension	mime	words	sentences	flesch	summary	cache	txt
work_pikiwx2qgndj5avsy435cqccsm	Jesse Dunietz	Automatically Tagging Constructions of Causation and Their Slot-Fillers	2017	18	.pdf	application/pdf	10313	984	62	describe two supervised approaches for tagging causal constructions and their arguments. outperform naı̈ve baselines for both construction recognition and cause and effect head Table 1: Examples of causal language, reflecting the annotation scheme described in §2.1 (with connectives in bold, the fundamental units of language are CONSTRUCTIONS – pairings of meanings with arbitrarily complex linguistic forms. related approaches for tagging causal constructions The scheme defines causal language as any construction which presents one event, state, action, or The connective annotation includes all words whose lemmas appear in Causeway is a system that performs this causal language tagging task. the cause and effect arguments of a causal instance. During training, the system examines each causal language instance, generating a TRegex pattern that will match tree fragments our causal connectives, and both tasks require identifying argument spans for each trigger. Annotating causal language using corpus lexicography of constructions.	./cache/work_pikiwx2qgndj5avsy435cqccsm.pdf	./txt/work_pikiwx2qgndj5avsy435cqccsm.txt