I enjoyed my research on Pure Mathematics, mainly guided by my PhD advisor Prof. Yuqi Guo.   However, in 2004, driven by my interest to solve practical problems, I decided to move to Computer Science. I joined the PhD programme in Computer Science amp; Engineering of CUHK where I obtained my second PhD (and later conducted months of postdoc research) under the supervision of Prof. Irwin King and Prof. Michael R. Lyu. In my thesis, I established three machine learning modelling frameworks: (i) Heat Diffusion Models on Random Graphs, (ii) Predictive Random Graph Ranking, and (iii) Random Graph Dependency. These belong to Statistical Modelling and Inference for Networks.  For my postdoctoral training I decided to work in Bioinformatics -- an area with many interesting practical problems which are both challenging and significant. During a period of more than three years with Prof. Alberto Paccanaro at Royal Holloway University of London, I have been developing novel graph-based statistical approaches (random walks, heat diffusion, semi-supervised machine learning, establishment of significance tests, etc.) for some significant biological problems such as inferring protein-protein interactions, calculating semantic similarity measures, predicting protein functions, and comparing different graphs. All these work belong to Bioinformatics using Statistical Methodology.   Now comes an ideal convergence point of my previous experiences: I am continuing my academic efforts on designingapplying statistical methods to tackle some Biological problems -- to which, I feel that, I am fully exploiting my abilities in Mathematics, Machine Learning, and Computer Science, and -- for which, my curiosity about living beings is substantially satisfied. A short phrase describing my current research is Bioinformatics amp; Statistical Modelling, especially of network data such as PPI, co-expression, and functional similarity.

My focus is in Bioinformatics amp; Statistical Modelling, especially of network data such as protein-protein interactions, co-expression, and functional similarity. A bio-molecular network can be viewed as a collection of nodes, representing the bio-molecules, connected by links, representing relations between the bio-molecules. I am working on inferring valuable information from bio-molecular networks.