L
Liang Du
Researcher at Shanxi University
Publications - 64
Citations - 1449
Liang Du is an academic researcher from Shanxi University. The author has contributed to research in topics: Cluster analysis & Feature selection. The author has an hindex of 18, co-authored 52 publications receiving 1030 citations. Previous affiliations of Liang Du include Chinese Academy of Sciences & Fudan University.
Papers
More filters
Proceedings ArticleDOI
Unsupervised Feature Selection with Adaptive Structure Learning
Liang Du,Yi-Dong Shen +1 more
TL;DR: In this paper, a unified learning framework is proposed to perform structure learning and feature selection simultaneously, where the structures are adaptively learned from the results of feature selection, and the informative features are reselected to preserve the refined structures of data.
Proceedings Article
Robust multiple kernel K-means using ℓ 2;1 -norm
TL;DR: A novel robust multiple kernel k-means algorithm that simultaneously finds the best clustering label, the cluster membership and the optimal combination of multiple kernels is proposed and an alternating iterative schema is developed to find the optimal value.
Posted Content
Unsupervised Feature Selection with Adaptive Structure Learning
Liang Du,Yi-Dong Shen +1 more
TL;DR: This work proposes a unified learning framework which performs structure learning and feature selection simultaneously, and demonstrates that the proposed method outperforms many state of the art unsupervised feature selection methods.
Proceedings ArticleDOI
Robust Spectral Learning for Unsupervised Feature Selection
Lei Shi,Liang Du,Yi-Dong Shen +2 more
TL;DR: A Robust Spectral learning framework for unsupervised Feature Selection (RSFS), which jointly improves the robustness of graph embedding and sparse spectral regression and robust Huber M-estimator is proposed.
Proceedings ArticleDOI
Robust Nonnegative Matrix Factorization via Half-Quadratic Minimization
Liang Du,Xuan Li,Yi-Dong Shen +2 more
TL;DR: A robust NMF method based on the correntropy induced metric, which is much more insensitive to outliers is proposed, and a half-quadratic optimization algorithm is developed to solve the proposed problem efficiently.