.png){kind=icon}
返回
Boosting classification tree-radial basis function network: Application in metabonomics studies
.png){kind=icon}
PRE
AI
DOI：10.1016/j.chemolab.2019.103829.png)
.png){kind=icon}
原文求助.png){kind=icon}
分享.png){kind=icon}
收藏
摘要
En 中文
In view of the growing complexity of metabolomics datasets, the existing chemometrics is increasingly faced with tremendous challenges. A central task of chemometrics for metabonomics data analysis lies in the informative variable selection, i.e., biomarker discovery. However, until now, limited attention has been paid to the reliability and robustness of biomarker discovery in metabonomics. In the current study, based on the intrinsic advantages of classification tree (CT) in variable selection as well as variable ranking and the properties of boosting in improving the reliability and robustness of a single model, boosting classification tree (BSTCT) was designed for robustly selecting the informative variables. Such an ensemble-based variable selection framework, i.e., BSTCT, iteratively constructed a set of CT classifiers on various weighted versions of the original training set. Each built CT providing the splitting variables and their corresponding rankings, the informative variables were successfully spied via inspecting the variable rankings over all CTs in BSTCT. The identified informative variables then acted as the inputs of RBFN, forming a new classification tool abbreviated as BSTCT-RBFN. A metabonomic dataset from the patients with lung cancer and the healthy controls was used as a case study to validate the performance of BSTCT-RBFN, with CT, support vector machine (SVM), multi-layer perception (MLP), and RBFN as comparisons. The results showed that BSTCT-RBFN can robustly and reliably find a shortlist of differentially expressed metabolites with statistical significance while attain more satisfactory classification accuracy than traditional CT, SVM, MLP, and RBFN.
Keyword:
Metabonomics
Chemometrics
Boosting classification tree
Radial basis function network
Boosting classification tree-radial basis function network
AI总结
对已上传原文的论文进行重点信息的提取，主要内容包括：简要概述、研究摘要、背景介绍、关键亮点、图文解析、展望与总结。
期刊
IF:
3.8
论文数:
4.6K
被引数:
1.2W
机构
引用论文
CO2 capture by piperazine mixed with non-aqueous solvent diethylene glycol in a rotating packed bed
PREAI
分享收藏Application of a Deep Neural Network to Metabolomics Studies and Its Performance in Determining Important Variables.png)
ANALYTICAL CHEMISTRY
ANALYTICAL CHEMISTRYIF6.7
PREAI
分享收藏
Particle swarm optimization-based protocol for partial least-squares discriminant analysis: Application to 1H nuclear magnetic resonance analysis of lung cancer metabonomics基于粒子群优化的偏最小二乘判别分析: 在肺癌代谢组学的1h核磁共振分析中的应用
PREAI
分享收藏分享收藏
查看更多内容