摘要
目的:伴随着机器学习方法的不断发展,在各个领域中得到了广泛的应用,将其应用到医学领域,成为当前的热点通过将机器学习方法应用到糖尿病预测中,以及提高当前的临床治疗效果。方法:通过使用机器学习算法来对糖尿病进行预测,使用微软的学习平台作为实验平台,采用逻辑回归神经网络决策树等,多种机器学习算法对糖尿病进行预测,对比多种算法的预测准确率。结果:通过实际研究发现使用决策树的临床预测效果最好,能够对糖尿病进行有效的预测。结论:通过应用决策树预测方法,能够有效地对糖尿病进行预测,在临床中具有较高的应用价值和推广意义,通过笔者的不断改进,使得整个预测结果正确率达到95.4%,因此在临床中具有较高的参考价值。
Objective:with the continuous development of machine learning methods,it has been widely used in various fields,and it has been applied to the medical field,and become a hot spot in the current field.It is necessary to apply machine learning method to diabetes prediction and improve the current clinical treatment effect.Methods:the prediction of diabetes was carried out by using machine learning algorithm,Microsoft learning platform was used as experimental platform,logic regression neural network decision tree,and many machine learning algorithms were used to predict diabetes,and the prediction accuracy of various algorithms was compared.Results:the results showed that the best clinical prediction effect was using decision tree,and it could predict diabetes effectively.Conclusion:by using decision tree prediction method,diabetes can be effectively predicted,which has a high application value and popularization significance in clinical.Through the continuous improvement of the author,the accuracy rate of the whole prediction results reaches 95.4%,so it has a high reference value in clinical.
作者
肖薇
XIAO Wei(Tianjin Polytechnic University,Tianjin 300000)
出处
《数字技术与应用》
2021年第4期104-106,共3页
Digital Technology & Application
关键词
机器学习算法
人工智能
糖尿病预测
Machine learning algorithm
Artificial intelligence
Diabetes prediction
