论文部分内容阅读
为快速、准确地对胎膜早破进行预测,首次应用了一种新型的数据挖掘技术-支持向量机预测模型.该模型针对所获取的胎膜早破及正常破膜数据集100个病例进行建模,并与神经网络、Logistic回归建模的性能进行了比较.结果表明,支持向量机具有可调参数少、学习速度快等优点,计算所得到的结果无论从准确率,还是所获取知识的可理解性等方面,都优于常用的神经网络等方法.用支持向量机方法建立的胎膜早破预测模型合理可行.
In order to predict the premature rupture of membranes quickly and accurately, a new data mining technique, Support Vector Machine (SVM) prediction model, is applied for the first time.The model is based on 100 cases of premature rupture of membranes and normal rupture of membrane data Modeling, and compared with the performance of neural network and Logistic regression model.The results show that the SVM has the advantages of less adjustable parameters, faster learning speed, etc. The results obtained from the calculation both in accuracy and acquired knowledge Are better than the commonly used methods such as neural network, etc. The prediction model of premature rupture of membranes with support vector machine is reasonable and feasible.