Bayesian Prediction Model Based on Attribute Weighting and Kernel Density Estimations

Table 3

Experimental results in terms of classifiers’ accuracy. Note that accuracies are estimated using 10-fold cross-validation with 95% confidence interval.

Data set	Naïve Bayes	AW-SKDE^MI	AW-LSKDE^MI
Anneal	93.99 ± 1.55	96.55 ± 1.19	76.17 ± 2.79
Balance-scale	91.36 ± 2.20	91.36 ± 2.20	89.6 ± 2.39
Breast-cancer	71.68 ± 5.22	72.38 ± 5.18	70.28 ± 5.30
Breast-w	97.28 ± 1.21	96.85 ± 1.29	88.41 ± 2.37
Colic	82.07 ± 3.92	81.79 ± 3.94	79.62 ± 4.12
Credit-a	85.94 ± 2.59	86.09 ± 2.58	83.62 ± 2.76
Dermatology	97.81 ± 1.50	97.81 ± 1.50	75.14 ± 4.43
Glass	77.10 ± 5.63	76.64 ± 5.67	62.62 ± 6.48
Heart-statlog	83.70 ± 4.58	83.70 ± 4.58	77.78 ± 5.15
Hepatitis	89.03 ± 4.92	89.03 ± 4.92	79.35 ± 6.37
Ionosphere	92.02 ± 2.83	91.45 ± 2.93	86.61 ± 3.56
Lymph	85.81 ± 5.62	85.81 ± 5.62	76.35 ± 6.85
Primary-tumor	50.15 ± 5.32	49.85 ± 5.32	24.78 ± 4.60
Segment	89.09 ± 1.27	88.70 ± 1.29	75.28 ± 1.76
Sick	97.48 ± 0.50	97.03 ± 0.54	93.88 ± 0.76
Vehicle	66.67 ± 3.18	66.90 ± 3.17	61.82 ± 3.27
Vote	90.11 ± 2.81	89.89 ± 2.83	91.49 ± 2.62
Average	84.78 ± 3.23	84.81 ± 3.22	76.05 ± 3.86