University of Guilan Journal of Mathematical Modeling 2345-394X 14 1 2026 03 01 A ‎robust ‎‎unsupervised ‎‎feature ‎s‎election based on ‎‎subspace ‎l‎earning and ‎‎adaptive ‎‎graph ‎‎structure 207 230 9129 10.22124/jmm.2025.30225.2714 EN Hazhir Sohrabi Department of Applied Mathematics‎, ‎University of Kurdistan‎, ‎Sanandaj‎, ‎Iran Shahrokh Esmaeili Department of Applied Mathematics, University of Kurdistan, Sanandaj, Iran Parham Moradi School of Engineering‎, ‎RMIT University‎, ‎Melbourne‎, ‎Australia Journal Article 2025 03 30 ‎Feature selection is vital for improving high-dimensional data analysis by identifying a subset of representative ‎and‎ ‎uncorrelated features‎.<br />‎This paper presents ‎an unsupervised feature selection ‎algorithm‎ based on subspace learning ‎and ‎adaptive ‎g‎raph ‎structure ‎(UFSAG‎)‎‎.<br />‎The ‎UFSAG uses matrix factorization to preserve global data structure and incorporates local correlations into its objective function‎.<br />‎It also integrates sample similarity graph learning to maintain data geometry‎.<br />‎Unlike prior methods‎, ‎UFSAG employs adaptive local structure learning to reduce noise and enhance feature selection‎.<br />‎By inducing row sparsity in the feature coefficient matrix using the $\ell_{2,1}$-norm‎, ‎UFSAG identifies representative features‎.<br />‎Comparative experiments on six datasets show UFSAG's superior clustering performance over twelve state-of-the-art methods‎. https://jmm.guilan.ac.ir/article_9129_51aea41a776790ee69c0f270bd847808.pdf