id	author	title	date	pages	extension	mime	words	sentences	flesch	summary	cache	txt
work_j365i7y6czfvfd2xgtshh7e7ei	Bob L. Sturm	The State of the Art Ten Years After a State of the Art: Future Research in Music Information Retrieval	2014	40	.pdf	application/pdf	21041	2358	73	information retrieval (MIR): the problem of music genre recognition (MGR). We perform a case study of all published research using the mostused benchmark dataset in MGR during the past decade: GTZAN. Of the 100 works using GTZAN, 96 employ the evaluation design Classify (an excerpt is Table 1: For each category of GTZAN: number of excerpts we identify by fingerprint (ENMFP); then searching manually (by self); number of songs tagged in last.fm (and number We now identify the excerpts in GTZAN, determine how music by specific artists Figure 4 shows how each GTZAN category is composed of music by particular artists. of an excerpt of music to one GTZAN category, we are interested here in a different question: GTZAN appears more than any other dataset in the evaluations of nearly 100 MGR publications (Sturm, 2012b). An analysis of the GTZAN music genre dataset.	./cache/work_j365i7y6czfvfd2xgtshh7e7ei.pdf	./txt/work_j365i7y6czfvfd2xgtshh7e7ei.txt