Not Logged In

An ensemble framework with l21-norm regularized hypergraph laplacian multi-label learning for clinical data prediction

Full Text: BHI2019.pdf PDF

—Previous work has shown that machine learning algorithms lend themselves to clinical decision-making and are a valuable tool for physicians. For clinical data, it is often necessary to assign multiple labels to a patient record by choosing from a large number of potential labels. A key problem in learning from multi-labelled data is how to exploit the information contained in the correlations between labels. Then hypergraph-based multi-label learning method learns from data by exploiting the spectral property of the hypergraph that encodes the correlation structure of labels. However, the problem with this method is the difficulty with which interpretations can be made. This is mainly due to its inability to recognize the importance of key features in the original feature space. Moreover, it is hard to comprehensively capture the complex structure of the correlations between labels. To overcome these difficulties and improve interpretability, we propose an l21-norm regularized Graph Laplacian multi-label learning to perform feature selection and label embedding simultaneously. In-depth experimental studies, using the publicly available Medical Information Mart for Intensive Care (MIMIC-III) database, validate the effectiveness of our approach.

Citation

P. Cao, Z. Teng, M. Huang, J. Yang, D. Zhao, A. Trabelsi, O. Zaiane. "An ensemble framework with l21-norm regularized hypergraph laplacian multi-label learning for clinical data prediction". International Workshop on Biomedical and Health Informatics, (ed: Illhoi Yoo, Jinbo Bi, Xiaohua Hu), pp 1436-1442, November 2019.

Keywords: Multi-label learning, High dimensionality, Feature selection, Ensemble classification, MIMIC-III
Category: In Workshop
Web Links: DOI
  IEEE

BibTeX

@misc{Cao+al:19,
  author = {Peng Cao and Zeyu Teng and Min Huang and Jinzhu Yang and Dazhe Zhao
    and Amine Trabelsi and Osmar R. Zaiane},
  title = {An ensemble framework with l21-norm regularized hypergraph laplacian
    multi-label learning for clinical data prediction},
  Booktitle = "Proceedings",
  Editor = {Illhoi Yoo, Jinbo Bi, Xiaohua Hu},
  Pages = {1436-1442},
  booktitle = {International Workshop on Biomedical and Health Informatics},
  year = 2019,
}

Last Updated: September 15, 2020
Submitted by Sabina P

University of Alberta Logo AICML Logo