论文部分内容阅读
针对兔胸部创伤后血液凝血功能检测数据,从变量检验入手,建立了基于Logistic回归模型的兔血液凝血功能诊断模型.结果表明模型的诊断正确率均在90%以上.为了降低样本集的划分对诊断正确率的影响,本文利用jackknife检验法思想和随机抽样技术,进一步对模型的有效性进行了评估分析并且证实了剔除纤维蛋白原(fibrinogen,FIB)指标会显著降低诊断正确率.
According to the data of blood coagulation test after thoracic trauma in rabbits, a logistic regression model based on Logistic regression model was established to diagnose the blood coagulation in rabbits. The results showed that the diagnostic accuracy of the model was over 90% .In order to reduce the classification of sample sets In this paper, we use the jackknife test method and random sampling technique to further evaluate the validity of the model and confirm that the removal of fibrinogen (FIB) index will significantly reduce the diagnostic accuracy.