.png){kind=icon}
.png){kind=icon}
Return
Multi-classification algorithm for compositional data based on Dirichlet feature embedding
.png){kind=icon}
PRE
AI
DOI：10.1080/03610918.2026.2645926.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
Compositional data are the data reflecting the relative information. The direct application of traditional classification algorithms to compositional data may yield misleading results. Existing methods are limited to either multi-classification with a single compositional feature or binary classification with multiple compositional features. This article proposes a novel algorithm for multi-classification with multiple compositional features, termed D-CoDAGSVM, which combines the Dirichlet feature embedding and directed acyclic graph support vector machine. To evaluate its performance, the proposed algorithm was compared with other methods in simulation studies and was also applied to a real metabolomics dataset. The results show that the proposed algorithm outperforms existing algorithms, confirming its effectiveness and usefulness.
Keywords:
Compositional data
Directed acyclic graph
Dirichlet distribution
Multi-classification
Support vector machine
Journal
C
IF:
0.8
Papers:
163
Citations:
4.7K