Table 4: The mean accuracy of classification from four classifiers based on two kinds of feature extraction.

SWNN (mean)RBF (mean)BP (mean)LS-SVM (mean)
cspW_DatacspW_ICcspW_DatacspW_ICcspW_DatacspW_ICcspW_DatacspW_IC

Player 187.1086.678.6185.282.7480.668.3772.0
Player 279.6682.972.1174.7275.9077.571.6468.0
Player 365.2974.083.6776.162.3772.867.2072.2
Player 476.4076.466.8167.5159.3171.271.5970.4
Player 560.8063.659.7253.9261.5463.358.2059.4
Player 674.6078.566.2777.254.8774.662.8167.5
Player 756.3076.349.5274.9772.1069.652.6160.1
Player 866.9481.349.8379.3053.3072.857.2262.0
Player 972.1377.4565.8173.6265.2677.363.7068.95
Player 1071.1683.650.682.057.075.159.7774.7
Mean71.0378.764.374.564.473.563.367.5
value0.0080.0420.0380.019

The classification results from four classifiers indicated that cspW_IC produced more quality features than cspW_Data. To investigate the statistical significance of the accuracies, we performed an analysis of variance (ANOVA) on each player’s result based on all classification accuracies (10 runs of the 10 × 10-fold cross-validation procedure). The -value from SWNN was 0.008, 0.042 from RBF neural network, 0.038 from BP neural network, and 0.019 from LS-SVM. These -values were leass than 0.05 for all players, which indicated that the difference was significant.