BARDBARD — Bayesian Argumentation via Delphi — uses causal Bayesian networks as underlying structured representations for argument analysis and automated Delphi methods to help groups of analysts develop, improve and present their analyses. This five-year project involves researching and designing new means of interacting with Bayesian networks, including new means of assessing their potential in causal explanations.

We are using causal Bayesian networks as the underlying structured representations for the argumentative/analytic domains of interest. Bayesian networks are ideal representations for systems involving uncertainty,  being well-established computational systems for probabilistic reasoning.

At Monash we’ve been working on Bayesian networks in Artificial Intelligence since the early 1990s, producing a wide range of models for applied sciences (e.g., environmental modeling, biosecurity, epidemiology) and tools for the automated learning of Bayesian networks from data.

In BARD we are designing and producing Graphical User Interfaces (GUIs) for using causal Bayesian networks as the underlying engines for arguments, allowing analysts to build and test competing or complementary arguments and to examine the impact of different pieces of evidence in an intuitive environment.

Delphi methods have been used for fifty years to help bring experts to improved opinions in domains of uncertainty, minimizing group think and other biases using anonymizing moderation.

At Monash we’ve developed automated support for the Delphi construction of Bayesian networks, which we will further enhance in BARD. BARD’s principal investigators includes experts in Delphi from the University of Strathclyde and experts in the psychology of causal reasoning from  Birkbeck College London and University College London.


Read more about: