Motif-Based Text Mining of Microbial Metagenome Redundancy Profiling Data for Disease Classification

Table 3

Classification results of pneumonia data in 2-class problem.

Method	Error rate		Dimension	Feature
	On training data	On test data
SVM/FMS	0.0922	0.1279	42	Microbes
SVM/PMF	0	0	551	Sequences
SVM/Kruskal-Wallis	0.01	0	28	Sequences
SVM/Information Gain	0	0.0116	26	Sequences
SVM/Chi-square statistic	0.01	0.0417	127	Sequences
SVM	0.0667	0.0116	4390	Sequences
NNA/FMS	0.1279	0.2393	20	Microbes
NNA/PMF	0	0.0833	361	Sequences
NNA/Kruskal-Wallis	0.0167	0.125	13	Sequences
NNA/Information Gain	0.0214	0.125	12	Sequences
NNA/Chi-square statistic	0.0381	0.125	26	Sequences
NNA	0.2667	0.5	4390	Sequences