id	author	title	date	pages	extension	mime	words	sentences	flesch	summary	cache	txt
work_wy64qplxp5ft7jjtc2f36mk7au	Fernando Fernández-Martínez	Text categorization methods for automatic estimation of verbal intelligence	2012	37	.pdf	application/pdf	11511	907	63	Test persons yielding a higher verbal intelligence (class HIGH ) may have test persons belonging to di"erent verbal intelligence classes may be distinguished by word or lemma patterns, even regardless of the order of these methods under test in terms of feature removal aggressiveness and classification accuracy improvement. keyword or feature set that reflects the most important words for all classes in kNN classification), the Rocchio approach relies on distance-based decisions (from a TC point of view, the relatedness of two documents can be Baseline Approach: Class-Based vs Corpus-Based Feature Selection Figure 6 presents the accuracy results obtained using either the corpusbased or the class-based feature selection methods. Figure 6: Baseline approach: Class-Based vs Corpus-Based feature selection. By confirming these di"erences with additional statistical evidence (i.e. more data), we may also conclude that the class-based feature selection improved the performance of the corpus-based one for the NB approach not	./cache/work_wy64qplxp5ft7jjtc2f36mk7au.pdf	./txt/work_wy64qplxp5ft7jjtc2f36mk7au.txt