摘要
为了更好地预测后天性脑损伤(ABI)患者认知功能康复的影响因素,借助于10折交叉验证测试算法,通过专一性、灵敏度和精度分析以及混淆矩阵分析对模型的性能进行测试,从而获得新的知识以评估和改善认知功能康复过程中的有效性。实验利用决策树(DT)、多层感知器(MLP)和广义回归神经网络(GRNN)三种预测模型对250例ABI案例进行了测试,结果表明,基于DT的模型的模拟结果明显比其他模型更为优越,预测平均精度可高达90.38%。
To better predicting factors of acquired brain injury （ABI） patients’ cognitive rehabilitation, prediction models based on Decision tree （DT）, multilayer perceptron （MLP） and general regression neural network （GRNN） is proposed.10-fold cross validation is carried out in order to test the algorithms .Specificity, sensitivity and accuracy analysis and confusion matrix anal-ysis are used to obtain new knowledge to evaluate and improve the effectiveness of the cognitive rehabilitation process .The experi-mental results show that results obtained by DT are clearly superior with a prediction average accuracy of 90 .38%.
出处
《东莞理工学院学报》
2013年第5期51-58,共8页
Journal of Dongguan University of Technology
关键词
后天性脑损伤
认知功能康复
数据挖掘
决策树
多层感知器
广义回归神经网络
acquired brain injury
cognitive rehabilitation
data mining
decision tree
multilayer perceptron
general regression neural network
