Title: A study on classification of small sample based on stochastic configuration networks

Authors: Saixian Yuan; Xuemei Yao; Yan Tang; Hongmei Zou

Addresses: Department of Statistics, Guizhou Minzu University, GuiYang, 520100, China ' School of Data Science and Information Engineering, Guizhou Minzu University, GuiYang, 520100, China ' Department of Statistics, Guizhou Minzu University, GuiYang, 520100, China ' Department of Statistics, Guizhou Minzu University, GuiYang, 520100, China

Abstract: Due to the limited availability of samples and the high cost of annotation, small sample classification presents significant challenges. Traditional models often struggle with poor generalisation and inadequate inter-class separability. To tackle these issues, this paper introduces an ensemble model called RS-SCN, which combines stochastic configuration networks (SCN) with the random subspace (RS) method. In this approach, the feature space is partitioned into random subspaces, each used to train an independent SCN model. The outputs of these models are then integrated through majority voting. This strategy reduces dependence on large-scale datasets while enhancing generalisation performance. Experimental results demonstrate that RS-SCN surpasses traditional methods in both generalisation and robustness. As a result, this approach enhances the applicability of SCN to both small-sample classification and function approximation tasks, offering an effective solution to the generalisation challenges posed by limited data.

Keywords: stochastic configuration networks; SCN; classification of small sample; random subspace method; subset of features; majority voting.

DOI: 10.1504/IJSCC.2026.153075

International Journal of Systems, Control and Communications, 2026 Vol.17 No.2, pp.107 - 123

Received: 12 Mar 2025
Accepted: 03 Jun 2025
Published online: 21 Apr 2026 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article