期刊文献+

基于迭代决策树的帕金森UPDRS预测模型研究 预览

Research on Parkinson UPDRS Prediction Model Based on GBDT
在线阅读 下载PDF
收藏 分享 导出
摘要 迭代决策树(GBDT)属于机器学习算法的一种,该算法具有较好的真实分布拟合能力,可用于解决大部分回归问题。根据帕金森病对不同年龄的男性和女性患者语音的影响不同这一现实依据,提出将性别和年龄这一先验知识融入到GBDT,实现对统一帕金森评定量表(UPDRS)的预测。将性别和年龄作为先验知识,对UPDRS预测模型进行模型分解;根据迭代决策树的原理,对分解后的各模型运用决策树进行模型重构,并在各自残差减少的梯度方向上迭代训练新的决策树;将得到的以叶子节点作为增益的决策树作为最终的UPDRS预测模型。在远程帕金森数据集的仿真实验中,得到的total-updrs和motor-updrs平均绝对误差值分别为4.4980和3.5318,与最小二乘法相比,分别提高了52.19%和53.36%,与决策树相比,分别提高了52.66%和52.89%。实验结果表明,根据先验知识,使用性别和年龄的组合进行预测模型分解,并对分解各模型分别进行模型重构,能够有效提高UPDRS预测的准确率。 As a kind of machine learning algorithm,gradient boosting decision tree(GBDT)can be used to solve most of the regression problem due to fine fitting ability of the true distribution.Based on the fact that the effect of Parkinson’s disease on the speech of male and female patients of different ages is different,we use the prior knowledge of gender and age into GBDT to predict unified Parkinson’s disease rating scale(UPDRS).Use sex and age as a prior knowledge to decompose the prediction model of UPDRS.Applying decision tree to reconstruct each new model and new decision tree is iteratively trained in the direction of the gradient of the respective residuals.Decision tree with leaf node as the gain is the final prediction model of UPDRS.In the simulation experiments of remote Parkinson data set,the mean absolute error(MAE)of total-updrs is 4.498 0 and the motor-updrs is 3.531 8,which are 52.19%and 53.36% higher than that of least squares method(LS),and 52.66%and 52.89%higher than that of the classification and regression tree(CART).The experiment show that GBDT based on sex and gender partition can improve the accuracy of UPDRS prediction.
作者 林钢 季薇 LIN Gang;JI Wei(School of Telecommunications and Information Engineering,Nanjing University of Posts and Telecommunications,Nanjing 210003,China)
出处 《计算机技术与发展》 2019年第1期216-220,共5页 Computer Technology and Development
基金 国家自然科学基金(61603197,61772284) 南京邮电大学科研基金(NY215104).
关键词 帕金森疾病 语音 统一帕金森评定量表 性别划分 年龄划分 迭代决策树 Parkinson's disease speech unified Parkinson's disease rating scale gender partition age partition gradient boosting decision tree
作者简介 林钢(1991-),男,硕士研究生,研究方向为机器学习;季薇,博士,副教授,研究方向为无线通信、机器学习。
  • 相关文献

参考文献6

二级参考文献97

  • 1孙斌 ,罗毅 .帕金森病的现代药物治疗[J].医药导报,2005,24(2):85-90. 被引量:7
  • 2伍业锋,赵彦云.北京地区人力资源竞争力状况及提升策略[J].经济师,2006(2):17-19. 被引量:5
  • 3王丽丽,苏德富.基于群体智能的选择性决策树分类器集成[J].计算机技术与发展,2006,16(12):55-57. 被引量:3
  • 4[1]Huglles AJ,Daniel SE,Kilford L,et al.Accuracy of the clinical diagnosis of idopathic Parkinson's disease:a clinico-pathological study of 100 cases[J].Neurol Neurosur Psychiat,1992,55:181-184. 被引量:1
  • 5[4]Delanty N,Dichter MA.Antioxidant therapy in neurologic disease[J].Arch Neurol,2000,57:1265-1270. 被引量:1
  • 6[5]Shults CW,Oakes D,Kieburtz K,et al.Effects of coenzyme Q10 in early Parkinson disease:evidence of slowing of the functional decline[J].Arch Neurol,2002,59:1541-1550. 被引量:1
  • 7[7]若山吉弘著.刘建武,等译.帕金森氏病防治[M].江西:江西科学技术出版社,2005,148-149. 被引量:1
  • 8Thompson S. Pruning boosted classifiers with a real valued genetic algorithm. Knowledge-Based Systems, 1999, 12(5-6): 277-284. 被引量:1
  • 9Zhou Z H, Tang W. Selective ensemble of decision trees// Proceedings of the 9th International Conference on Rough Sets, Fuzzy Sets, Data Mining, and Granular Computing. Chongqing, China, 2003:476-483. 被引量:1
  • 10Hernandez-Lobato D, Hernandez-Lobato J M, Ruiz-Torrubiano R, Valle A. Pruning adaptive boosting ensembles by means of a genetic algorithm//Corchado et al. International Conference on Intelligent Data Engineering and Automated Learning. Berlin Heidelberg: Springer-Verlag, 2006: 322- 329. 被引量:1

共引文献108

投稿分析

相关作者

内容加载中请稍等...

相关机构

内容加载中请稍等...

相关主题

内容加载中请稍等...

浏览历史

内容加载中请稍等...
;
使用帮助 返回顶部 意见反馈