School of
Information Technology and Electrical Engineering

Speaker: Sitthichoke Subpaiboonkit
Seminar Date: Wed, 29/11/2017 - 10:00
Venue: 78-631; Data Science MMLab
Host: Prof Xue Li

Seminar Type:  PhD Confirmation Seminar


Causality involves with most of knowledge disciplines, like medicine, pharmacology, biology, economics and social science. It can be described as the connections between the causes and effect events. Causality discovery leads to the understanding of the mechanism of how events happen. Causality discovery is a crucial task. Generally, it can be performed by a gold standard method called randomised controlled trail. However, it is impractical in many real-world cases because of cost, ethics and time. Causal Bayesian Networks is the prominent computational method used to discovery causality in observational data. Constraint-base causal discovery approach becomes an alternative way by considering only sub-graph of whole Bayesian Network. However, both of these methods are high computational.

In this study, I identified the three problems on causality discovery. The First problem is to discover causal relation from drug-drug interaction to adverse effect. Drugs consumed together may cause undesired effects. Discovery this causality can help to reduce loss in the healthcare. I proposed a novel method called Domain Knowledge-Centred (DKC), which uses a known cause and effect to discovery other causes of the effect. For the second problem, I will investigate the causality discovery from drug-drug interaction to adverse effect in sub-population. This can provide more precise suggestions to doctors to prevent adverse effect. Lastly, I plan to solve the gene regulatory network problem, which is also a causality discovery problem.


Sitthichoke Subpaiboonkit: received his B.S Degree in Computer Science in 2007, and M.S Degree in Bioinformatics in 2011 from Chiang Mai University, Thailand. Currently, he is doing PhD study in at the School of Information Technology and Electrical Engineering, University of Queensland, under the supervision of Prof Xue Li. His research interests include machine learning, causality discovery and bioinformatics.