M
Michael K. Ng
Researcher at University of Hong Kong
Publications - 658
Citations - 24376
Michael K. Ng is an academic researcher from University of Hong Kong. The author has contributed to research in topics: Cluster analysis & Computer science. The author has an hindex of 72, co-authored 608 publications receiving 20492 citations. Previous affiliations of Michael K. Ng include The Chinese University of Hong Kong & Vanderbilt University.
Papers
More filters
Journal ArticleDOI
Zoom-based super-resolution reconstruction approach using prior total variation
TL;DR: A robust and efficient approach for zoom-based super-resolution (SR) reconstruction problems, which employs the total variation of the desired image priori in the maximum a-posteriori estimation.
Journal ArticleDOI
Collective prediction of protein functions from protein-protein interaction networks.
TL;DR: This paper proposes an effective Markov chain based CC algorithm (ICAM) to tackle the label deficiency problem in CC for interrelated proteins from PPI networks, and uses an ICA framework to iteratively refine the learning models for improving performance of protein function prediction from P PI networks in the paucity of labeled data.
Journal ArticleDOI
Multi-Label Classification by Semi-Supervised Singular Value Decomposition
TL;DR: This paper proposed to use a semi-supervised singular value decomposition (SVD) to handle multi-label problems, and takes advantage of the nuclear norm regularization on the SVD to effectively capture the label correlations.
Journal ArticleDOI
Kernel Density Estimation Based Multiphase Fuzzy Region Competition Method for Texture Image Segmentation
Fang Li,Michael K. Ng +1 more
TL;DR: This paper proposes a multi-phase fuzzy region competition model for texture image segmentation and shows that the proposed method is competitive with the other state-of-the-art segmentation methods.
Journal ArticleDOI
Dictionary Learning-Based Subspace Structure Identification in Spectral Clustering
TL;DR: The main contribution of this paper is to consider both nonnegativity and sparsity constraints together in DL such that data can be represented effectively by nonnegative and sparse coding coefficients and nonnegative dictionary bases.