What question did this study set out to answer?

To develop and validate a stable multi-objective particle swarm optimization algorithm capable of handling high-dimensional, small-sample metabolomics data.

August 19, 2026MetabolitesOpen Access

MPR-MOPSO-ASFS: A Stable Multi-Objective Feature Selection Algorithm for Metabolomics

View Full Paper

Ask AI

Bookmark

Authors

QYQing YeJiangxi University of Traditional Chinese Medicine 邓邓增Jiangxi University of Traditional Chinese Medicine XXXin XieJiangxi University of Traditional Chinese Medicine

Discussion

Member takes

Implication

Computational study demonstrates improved feature selection accuracy and stability in high-dimensional metabolomics datasets, suggesting enhanced biomarker discovery for biomedical research.

Key Points

To develop and validate a stable multi-objective particle swarm optimization algorithm capable of handling high-dimensional, small-sample metabolomics data.
Constructed a stability-driven bi-objective optimization model incorporating the harmonic mean of classification accuracy and feature subset stability.
Integrated a Maximum Pattern Recognition (MPR) filter for initial rapid dimensionality reduction alongside an Adaptive Sparsity Feature Selection (ASFS) mechanism.
Evaluated performance across three metabolomics datasets and five public high-dimensional, small-sample datasets against benchmark algorithms including CMDPSOFS and MOEAD-FS.
The MPR-MOPSO-ASFS algorithm demonstrated superior Pareto front quality and higher classification accuracy compared to baseline algorithms.
The combination of the MPR filtering stage and adaptive sparsity optimization significantly enhanced algorithm stability and compactness of selected feature subsets.

Cite This Study

Ye et al. (2026) studied this question.

synapsesocial.com/papers/6a85638803308d306e2d6b33 https://doi.org/10.3390/metabo16080581

View Full Paper

Ask AI

Bookmark

Implication

Computational study demonstrates improved feature selection accuracy and stability in high-dimensional metabolomics datasets, suggesting enhanced biomarker discovery for biomedical research.

MPR-MOPSO-ASFS: A Stable Multi-Objective Feature Selection Algorithm for Metabolomics

Discussion

Member takes

Implication

Key Points

Cite This Study

MPR-MOPSO-ASFS: A Stable Multi-Objective Feature Selection Algorithm for Metabolomics

Discussion

Member takes

Implication

Key Points

Cite This Study

Also Consider

Also Consider