id	author	title	date	pages	extension	mime	words	sentence	flesch	summary	cache	txt
cana-4843	Vaka Padmavathi	Investigate Effect of Various Feature Selection Methods for Attack Detection of DOH Traffic	2025	17	.pdf	application/pdf	5651	260	42	Communications on Applied Nonlinear Analysis ISSN: 1074-133X Vol 32 No. 10s (2025) 499 https://internationalpubls.com Table 3: Performance Measures for Model-2(B-M) Features # Accuracy Precision Recall F1 ROC AUC 10,15, 0.997 0.998 0.999 0.998 0.987 20,25,30 552 161 199 680 939 16,18, 0.99 0.99 0.999 0.998 0.987 21,22,24 775 814 439 789 995 13,15, 0.997 0.998 0.999 0.998 0.988 16,17,18 737 304 261 782 545 Figure 3: Feature based performance of Model2 Figure: 4 PCA based ROC Curve for two models as Shown in Table 5. The k-NN algorithm's performance is evaluated before and after feature selection to ensure consistency and reliability in detecting malicious DoH connections.	cache/cana-4843.pdf	txt/cana-4843.txt