id	author	title	date	pages	extension	mime	words	sentences	flesch	summary	cache	txt
work_kqyfvfw3crfzvdcucr2np6fbha	Carmen Klaussner	Finding characteristic features in stylometric analysis	2015.0	22	.pdf	application/pdf	9661	765	67	2002) was designed for authorship attribution, seeking the most likely authorial candidate for a given document from a set of authors based on differences between z-scores feature frequencies are consistent over the target author's document set. variable features in the language, but will score highly those features whose use contrasts between the target author's documents and the reference set. Collins in a two-author comparison and a larger reference set comprising different namely a set consisting of twenty-four texts by Dickens and Collins each (shown in the number of highest features selected from Dickens' and the reference corpus' representative Instead, we also consider the same number of most prominent representative and distinctive features and compare how many items are shared, when the same For the comparison between the CoV Tuning method (Hoover, 2014) and Representativeness and Distinctiveness, we again consider the Dickens/Collins data set.	./cache/work_kqyfvfw3crfzvdcucr2np6fbha.pdf	./txt/work_kqyfvfw3crfzvdcucr2np6fbha.txt