Identification of host-microbe interaction factors in the genomes of soft rot-associated pathogens Dickeya dadantii 3937 and Pectobacterium carotovorum WPP14 with supervised machine learning

Table 4 Statistics for positive class object prediction and parameters used in selected learning schemes for both Dickeya dadantii 3937 and Pectobacterium carotovorum WPP14

Classifiers	Precision	TPR/recall/sensitivity	specificity/TNR	accuracy	F-measure	AUC
Dd3937
Random Forest	0.93	0.81	0.98	0.94	0.87	0.97
Bayesian Network	0.91	0.85	0.97	0.94	0.88	0.97
SMO using RBF kernels	0.93	0.85	0.98	0.95	0.89	0.92
SMO using polynormial kernels	0.91	0.87	0.97	0.95	0.95	0.89
Adaptive Boosting (Naïve Bayes)*	0.84	0.89	0.95	0.93	0.87	0.96
Adaptive Boosting (Decision Tree)*	0.96	0.91	0.99	0.97	0.93	0.98
Adaptive Boosting (IBK)*	0.96	0.84	0.99	0.95	0.90	0.99
Adaptive Boosting (Decision Stump)*	0.92	0.87	0.98	0.95	0.89	0.97
Multi-Boosting (Decision Tree)*	0.97	0.91	0.99	0.97	0.94	0.98
Multi-Boosting (IBK)*	0.91	0.77	0.98	0.93	0.84	0.93
Multi-Boosting (Naïve Bayes)*	0.90	0.91	0.97	0.95	0.91	0.96
Logit-Boosting (Decision Stump)*	0.91	0.90	0.97	0.96	0.91	0.98
WPP14
Random Forest	0.89	0.81	0.97	0.93	0.85	0.97
Bayesian Network	0.90	0.83	0.97	0.94	0.87	0.97
SMO using RBF kernels	0.94	0.84	0.98	0.95	0.89	0.91
SMO using polynormial kernels	0.93	0.86	0.98	0.95	0.95	0.89
Adaptive Boosting (Naïve Bayes)*	0.89	0.89	0.97	0.95	0.89	0.96
Adaptive Boosting (Decision Tree)*	0.95	0.86	0.99	0.96	0.90	0.98
Adaptive Boosting (IBK)*	0.87	0.83	0.96	0.93	0.85	0.92
Logit-Boosting (Decision Stump)*	0.90	0.85	0.97	0.94	0.88	0.97
Multi-Boosting (Decision Tree)*	0.94	0.86	0.98	0.96	0.90	0.98
Multi-Boosting (Decision Stump)*	0.91	0.75	0.98	0.93	0.82	0.97
Multi-Boosting (Naïve Bayes)*	0.90	0.89	0.97	0.95	0.89	0.96
Logit-Boosting (Decision Stump)*	0.90	0.87	0.97	0.95	0.89	0.97

*: denote ensemble classifiers, with base learner being shown within parenthesis.
Abbr: SMO: Support Vector Machine using Sequential Minimal Optimization; IBK: instance based learner with K-nearest neighbor classifier; RBF: Radial Basis Function kernels.

ISSN: 1471-2164