We have keen interests in both statistical methodology development (big data analytics, experimental design, machine learning) and real data applications (biomedical, educational, financial). The multi-year experience with financial and educational industries made a gap in Dr. Zhang’s publication record, but filled his mind with a great deal of real data sense. The research works listed below are all based on the new projects after Dr. Zhang joined HKU in 2016, while those old works (before 2009) can be referred here.

Big Data Analytics

The rapid growth of sample size and feature dimensionality poses challenges for traditional statistical methods. To tackle such big data challenges, we study the effective ways for performing statistical modeling based on reduced data via subsampling, projection, partitioning and aggregation techniques.

Structural Filtering

Structural assumptions like sparsity and piecewise smoothness are often made when fitting a statistical model in high dimensional space. These structural properties can be studied by L0-regularization in parametric and nonparametric settings. In this project, we aim at developing efficient algorithms for solving such structural filtering problems.

R:BeSS package
R:AMIAS package

Experimental Design

UniDOE: a stochastic optimization package for constructing uniform space-filling designs

R:UniDOE package

Medical Image Analysis

Fully automated deep learning approach vs. conventional radiomics models.

2018 Data Science Bowl (Top 1%, Silver solution)

Financial Risk Modeling

Dual-time analytics and joint age-horizon credit risk models.

Educational Data Mining

Machine learning approach vs. traditional latent variable modeling in large-scale educational assessments as well as learning anlaytics.