id	author	title	date	pages	extension	mime	words	sentences	flesch	summary	cache	txt
work_7hmcw6dxwjborfxvzwffuce4wq	Ivica Slavkov	Error curves for evaluating the quality of feature rankings	2020	39	.pdf	application/pdf	27802	5502	93	In this article, we propose a method for evaluating feature ranking algorithms. (RFA) curves, which reveal how the relevant features are distributed in the ranking(s). (true positives) and irrelevant features (false positives) with the feature relevance above the starting point for showing the usefulness of a feature ranking evaluation method, as why FFA curves are an appropriate method for comparing feature rankings, nor which and lower quality of feature rankings is reflected in the FFA and RFA curves, and thus When comparing the FFA and RFA curves of different ranking methods, constructed with We construct the curves that base on the feature ranking methods described in Figure 7 Ranking quality assessment for datasets breast-w (A–C) and water (D–F) in terms of the FFA (A and D) and RFA curves (B and E), Error curves for evaluating the quality of feature rankings Error curves for evaluating the quality of feature rankings	./cache/work_7hmcw6dxwjborfxvzwffuce4wq.pdf	./txt/work_7hmcw6dxwjborfxvzwffuce4wq.txt