• 主题
高级搜索  >
历史搜索 清空历史
首页> 外文会议>International conference on artificial intelligence;ICAI 2011 >Prediction of Chronic Fatigue Syndrome Using Decision Tree-Based Ensemble Methods
【24h】

Prediction of Chronic Fatigue Syndrome Using Decision Tree-Based Ensemble Methods

机译：基于决策树的集成方法对慢性疲劳综合症的预测

获取原文
页面导航
  • 摘要
  • 著录项
  • 相似文献
  • 相关主题

摘要

A decision tree ensemble classifier is a machine learning technique in which several decision trees are joined in an effort to improve the classification accuracy of a prediction model. In this paper, a three-level decision tree ensemble is proposed in the classification of chronic fatigue syndrome (CFS). While existing CFS classification models are based largely on biomarker data, immunological data, and other biologically-based datasets, we are proposing to create a classification model based on patient responses to a medical survey. We found that decision tree ensemble approaches generally performed poorly in comparison to single decision tree methods. This poor performance may be explained by an accumulation of error through the levels of the ensemble, a variable-sample imbalance, the large amount of missing values present in the dataset, the use of a subjective method of data retrieval (self-report), or any combination of these factors.
机译：决策树集成分类器是一种机器学习技术，其中为了提高预测模型的分类准确性而将多个决策树结合在一起。在慢性疲劳综合症（CFS）的分类中，本文提出了一种三级决策树集合。尽管现有的CFS分类模型主要基于生物标记数据，免疫学数据和其他基于生物学的数据集，但我们建议根据患者对医学调查的反应来创建分类模型。我们发现，与单决策树方法相比，决策树集成方法通常效果较差。表现不佳的原因可能是：由于整体水平的误差累积，样本可变变量不平衡，数据集中存在大量缺失值，使用数据检索的主观方法（自我报告），或这些因素的任意组合。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱：kefu@zhangqiaokeyan.com

京公网安备：11010802029741号 ICP备案号：京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有

  • 客服微信

  • 服务号