Skip to main content

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