id	author	title	date	pages	extension	mime	words	sentences	flesch	summary	cache	txt
work_p2gfo6whijbf3nveq7bcobpptm	Swapna Somasundaran	Towards Evaluating Narrative Quality In Student Writing	2018	16	.pdf	application/pdf	10295	958	61	Automated scoring of narrative essays is a challenging area, and one that has not been explored extensively in NLP research. We investigated the effectiveness of each feature for scoring narrative traits the first detailed annotation study on scoring narrative essays for different aspects of narrative quality. (2) We present an automated system for scoring narrative quality, with linguistic features specific to encoding aspects of good story-telling. for evaluating student writing, or whether encoding features that capture different narrative aspects The rubric provides the following criteria for an essay of score point 4 in terms of five aspects or sub-traits: "The organization of the narrative E-rater (Attali and Burstein, 2006), a state-of-theart commercial system for automatic essay scoring, uses a comprehensive suite of features covering many aspects of writing quality, such as grammar, language use, mechanics, fluency, style, organization, and development. Table 4: Performance (QWK) on predicting traits and Narrative and Total scores; Best feature combinations:	./cache/work_p2gfo6whijbf3nveq7bcobpptm.pdf	./txt/work_p2gfo6whijbf3nveq7bcobpptm.txt