Improving statistical learning methods via features selection without replacement sampling and random projection

2025-05-28

AI Summary

To address overfitting and low classification accuracy arising from the “small-sample, high-dimensional” nature of glioblastoma microarray data (GSE50161), this paper proposes a novel ensemble classification framework integrating Kendall’s rank correlation test, Feature Selection without Replacement (FSWOR), and Linear Discriminant Analysis (LDA) projection. For the first time, FSWOR is synergistically combined with LDA to jointly achieve dimensionality reduction and discriminative feature enhancement, while Kendall’s test is employed to select biologically significant genes. Under k-fold cross-validation, the model achieves 96.0% classification accuracy—improving upon state-of-the-art methods by 9.09%. The feature dimensionality is reduced from 54,675 to 20,890, substantially mitigating overfitting and enhancing both model robustness and biological interpretability. This work introduces a methodological innovation in microarray analysis by unifying feature selection, discriminative projection, and ensemble learning into a coherent pipeline.

Technology Category

• Machine Learning: Dimensionality Reduction/Feature Selection
• Search and Optimization: Mixed Discrete/Continuous Search
• Natural Language Processing: Interpretability, Analysis, and Evaluation of NLP Models

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Graph Algorithms and Modeling for the Web: Graph embeddings and representation learning for Web-related graphs
• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking

Abstract

Cancer is fundamentally a genetic disease characterized by genetic and epigenetic alterations that disrupt normal gene expression, leading to uncontrolled cell growth and metastasis. High-dimensional microarray datasets pose challenges for classification models due to the"small n, large p"problem, resulting in overfitting. This study makes three different key contributions: 1) we propose a machine learning-based approach integrating the Feature Selection Without Re-placement (FSWOR) technique and a projection method to improve classification accuracy. 2) We apply the Kendall statistical test to identify the most significant genes from the brain cancer mi-croarray dataset (GSE50161), reducing the feature space from 54,675 to 20,890 genes.3) we apply machine learning models using k-fold cross validation techniques in which our model incorpo-rates ensemble classifiers with LDA projection and Na""ive Bayes, achieving a test score of 96%, outperforming existing methods by 9.09%. The results demonstrate the effectiveness of our ap-proach in high-dimensional gene expression analysis, improving classification accuracy while mitigating overfitting. This study contributes to cancer biomarker discovery, offering a robust computational method for analyzing microarray data.

Problem

• Addresses overfitting in high-dimensional microarray cancer datasets
• Identifies significant genes via feature selection and statistical tests
• Improves classification accuracy using ensemble methods and projections

Innovation

• Feature Selection Without Replacement sampling technique
• Kendall test for significant gene identification
• Ensemble classifiers with LDA projection