Heart rate monitoring using human speech spectral features

James, Alex Pappachen

doi:10.1186/s13673-015-0052-z

Table 1 Comparison of different classifiers when 5 % of the class samples are used as training data and remaining 95 % of sample are used as test on emotion database

From: Heart rate monitoring using human speech spectral features

Method	% Correct	AUC	F-measure	Precision (%)	Recall (%)
BayesNet [14]	34.86 ± 2.37	0.52 ± 0.03	0.45 ± 0.12	35 ± 0.09	79 ± 0.34
ComplementaryBayes [15]	39.05 ± 0.88	0.57 ± 0.02	0.45 ± 0.11	37 ± 0.09	56 ± 0.14
NaiveBayes [10]	40.09 ± 1.12	0.60 ± 0.02	0.43 ± 0.04	42 ± 0.02	46 ± 0.08
NaiveBayesMultinomial [16]	38.62 ± 0.59	0.61 ± 0.00	0.45 ± 0.07	40 ± 0.01	53 ± 0.10
Logistic [17]	40.83 ± 1.17	0.61 ± 0.01	0.44 ± 0.05	43 ± 0.02	46 ± 0.07
MultilayerPerceptron [18]	40.27 ± 1.39	0.61 ± 0.02	0.41 ± 0.10	42 ± 0.10	43 ± 0.17
SimpleLogistic [17]	40.63 ± 1.31	0.61 ± 0.01	0.43 ± 0.07	42 ± 0.06	46 ± 0.09
IB1 [19]	37.17 ± 0.90	0.53 ± 0.01	0.37 ± 0.02	38 ± 0.01	37 ± 0.03
Ibk [19]	37.21 ± 0.84	0.54 ± 0.01	0.40 ± 0.01	37 ± 0.01	42 ± 0.02
Kstar [20]	40.11 ± 1.18	0.59 ± 0.02	0.41 ± 0.03	41 ± 0.02	42 ± 0.05
LWL [21]	39.37 ± 1.22	0.59 ± 0.02	0.40 ± 0.09	43 ± 0.05	42 ± 0.17
Bagging [13]	39.54 ± 1.09	0.58 ± 0.02	0.41 ± 0.02	41 ± 0.02	41 ± 0.04
ClassificationViaClustering [22]	37.19 ± 2.31	0.53 ± 0.03	0.29 ± 0.21	25 ± 0.18	35 ± 0.25
ClassificationViaRegression [23]	40.47 ± 1.33	0.60 ± 0.01	0.42 ± 0.08	42 ± 0.07	44 ± 0.13
CVParameterSelection [11]	33.33 ± 0.00	0.50 ± 0.00	0.50 ± 0.00	33 ± 0.00	100 ± 0.00
FilteredClassifier [12]	34.86 ± 2.37	0.52 ± 0.03	0.45 ± 0.12	35 ± 0.09	79 ± 0.34
DTNB [24]	36.62 ± 2.41	0.54 ± 0.03	0.36 ± 0.16	38 ± 0.14	50 ± 0.37
JRip [25]	36.82 ± 1.95	0.53 ± 0.03	0.29 ± 0.17	37 ± 0.18	36 ± 0.35
J48 [13]	39.24 ± 1.37	0.56 ± 0.02	0.37 ± 0.11	43 ± 0.08	38 ± 0.18
NBTree [26]	38.89 ± 1.18	0.56 ± 0.02	0.35 ± 0.12	45 ± 0.09	36 ± 0.20

Back to article page