Meet the Engagement Team – Bayesian Network Modelling Association

Bezalem Eshetu Yirdaw	Country, city, institution/position	PhD student, University of South Africa (currently residing in Ethiopia)
	Field of research or work	Multilevel Bayesian Network
	Why using BNs/DAGs	Bezalem uses Bayesian Networks because they provide a transparent and flexible probabilistic framework for modelling complex dependencies among multiple outcomes. Incorporating hierarchical and repeated-measure patterns allows the model to capture real-world variability due to clustering, producing outputs that are interpretable and actionable for decision making.
Bruce Marcot	Country, city, institution/position	USA, Oregon, PhD Emeritus Research Wildlife Biologist, U.S. Forest Service, Pacific Northwest Research Station
	Field of research or work	Integration of BN modelling with decision science, expert elicitation, rare event statistics, and environmental monitoring
	Why using BNs/DAGs	Bruce has applied Bayesian Networks and related modelling to major environmental and conservation issues to support decision- and policy-making across regional, domestic, and international contexts, including old-forest management, rare and little-known species analysis, and biodiversity assessment.
Kate Parkins	Country, city, institution/position	Australia, Melbourne, The University of Melbourne, Research Fellow
	Field of research or work	Fire risk modelling, fire ecology
	Why using BNs/DAGs	Decision-making is a major part of fire risk research, and trade-offs often need to be considered across people, property, and the environment. Bayesian Networks are a valuable way to explore these trade-offs and help guide decision making.
Kieran Drury	Country, city, institution/position	PhD Student, University of Warwick, UK
	Field of research or work	Elicitation of Bayesian Networks to support decision making, applied to criminal proceedings and environmental modelling
	Why using BNs/DAGs	Kieran sees Bayesian Network models as one of the most accessible yet flexible types of probabilistic model. This is especially important in decision problems, where decision makers need to understand and trust the model in order to have confidence in its outputs.
Mari Takashima	Country, city, institution/position	Australia, Brisbane, University of Queensland, Research Fellow
	Field of research or work	Clinical trials, health service research
	Why using BNs/DAGs	Mari’s work focuses on understanding causal mechanisms, and BN or DAG approaches help map these clearly. This supports the responsible identification of risks, highlights where preventable harm may arise, and strengthens the safety and impact of healthcare research.
Minh Vo	Country, city, institution/position	Australia, Melbourne, Graduated Master of AI – University of Monash, Lead Agentic AI Engineer – Grant Help
	Field of research or work	Real-time explaining Bayesian Networks using large language model
	Why using BNs/DAGs	Minh explores the use of Large Language Models with structured prompting to automate the causal explanations needed to keep Bayesian Networks transparent and useful as models become larger and more sophisticated.
Steven Mascaro	Country, city, institution/position	Australia, Melbourne, Monash University, Faculty of IT, Senior Lecturer, Bayesian Intelligence, Director and Senior Consultant
	Field of research or work	Bayesian network methodology, elicitation, explainability, machine learning, causal inference and discovery
	Why using BNs/DAGs	Steven views Bayesian Networks as powerful and faithful representations of causal reasoning under uncertainty, with strong potential for safe and understandable AI in high-stakes decision making. His work aims to make Bayesian Networks accessible, scalable, transparent, and practical for a wide range of applications.