BMC Bioinformatics

Table 2 Comparison of performance of M5Rules and RNN models on different 3 × 1 dataset combinations

From: Predicting rice blast disease: machine learning versus process-based models

Training data*	Test data	Train		Test
Training data*	Test data	r (a)	r (b)	r (a)	r (b)	MAE** (a)	MAE** (b)	r² (a)	r² (b)
k2015 + s2016 + p2015	k2016	0.94	0.96	0.76	0.81	0.52	0.49	0.58	0.66
k2016 + s2016 + p2015	k2015	0.89	0.91	0.29	0.72	0.49	0.58	0.09	0.52
k2016 + k2015 + p2015	s2016	0.88	0.86	0.77	0.75	0.88	0.95	0.60	0.56
k2016 + k2015 + s2016	p2015	0.91	0.87	0.57	0.53	0.62	0.99	0.32	0.28
Average Values				0.60	0.70	0.63	0.75	0.40	0.51

*k = Kalochori, s = Seville, p = Portugal; **Mean absolute error, (a) = M5Rules, (b) = RNN

Back to article page

ISSN: 1471-2105

Contact us

General enquiries: journalsubmissions@springernature.com