id	author	title	date	pages	extension	mime	words	sentences	flesch	summary	cache	txt
work_duq33giwkjd7djgxwuk3nmjagm	Dat Ba Nguyen	J-NERD: Joint Named Entity Recognition and Disambiguation with Rich Linguistic Features	2016	13	.pdf	application/pdf	8579	847	71	Methods for Named Entity Recognition and Disambiguation (NERD) perform Motivation: Methods for Named Entity Recognition and Disambiguation, NERD for short, typically proceed in two stages: • At the NER stage, text spans of entity mentions are detected and tagged with coarsegrained types like person, organization, location, etc. not able to help NER, for example, by disambiguation "easy" mentions (e.g., of prominent entities probabilistic graphical model for the joint recognition and disambiguation of named-entity mentions • an inference method that maintains the uncertainty of both mention candidates (i.e., token spans) and entity candidates for competing entity candidates; 2) linguistic features about verbal phrases from dependency parse trees; 3) maintaining candidates for both mentions and entities entity, for a given mention token toki, based on the NER type and the NED label of a token toki generates binary features if toki appears in uppercase form and is NER-labeled as typej in the training corpus.	./cache/work_duq33giwkjd7djgxwuk3nmjagm.pdf	./txt/work_duq33giwkjd7djgxwuk3nmjagm.txt