id	author	title	date	pages	extension	mime	words	sentences	flesch	summary	cache	txt
work_6ca2tg546ratlmrt7cjdctbwau	Lea Frermann	Whodunnit? Crime Drama as a Case for Natural Language Understanding	2018	16	.pdf	application/pdf	8782	894	63	dataset1 based on CSI episodes, formalize perpetrator identification as a sequence labeling We formalize the task of identifying the perpetrator in a crime series as a sequence labeling problem. The model predicts for each input whether the perpetrator is mentioned or not. work in natural language processing, computer vision, and more generally multi-modal learning. work uses low-level features (e.g., based on face detection) to establish social networks of main characters in order to summarize movies or perform genre Once the episode is finished and the perpetrator is revealed, the same participant annotates entities in the screenplay referring We formalize the problem of identifying the perpetrator in a crime series episode as a sequence labeling task. As mentioned earlier, the input to our model consists of a sequence of sentences, either spoken utterances or scene descriptions (we do not use speaker model guesses at the start of the episode closely follow the pattern of gold-perpetrator mentions (bottom	./cache/work_6ca2tg546ratlmrt7cjdctbwau.pdf	./txt/work_6ca2tg546ratlmrt7cjdctbwau.txt