How to collaboratively build Bayesian networks in BARD
How to collaboratively build Bayesian networks in BARD
In many complex, real-world situations, problem solving and decision making require effective reasoning about causation and uncertainty. However, human reasoning in these cases is prone to confusion and error. Bayesian networks (BNs) are an artificial intelligence technology that models uncertain situations, supporting probabilistic and causal reasoning and decision making. However, to date, BN methodologies and software require significant upfront training, do not provide much guidance on the model building process, and do not support collaboratively building BNs. BARD (Bayesian ARgumentation via Delphi) is both a methodology and an application that utilizes (1) BNs as the underlying structured representations for better argument analysis, (2) a multi-user web-based software platform and Delphi-style social processes to assist with collaboration, and (3) short, high-quality e-courses on demand, a highly structured process to guide BN construction, and a variety of helpful tools to assist in building and reasoning with BNs, including an automated explanation tool to assist effective report writing. The result is an end-to-end online platform, with associated online training, for groups without prior BN expertise to understand and analyze a problem, build a model of its underlying probabilistic causal structure, validate and reason with the causal model, and use it to produce a written analytic report. Initial experimental results demonstrate that BARD aids in problem solving, reasoning and collaboration.
In this workshop, participants will be given a live demo of the BARD software (with some Q&A along the way), then participants will be put in BARD groups to collaboratively build a small example BN. Attendees will be given a temporary demo account to use during the webinar (and for one month afterwards), that will also provide them with access to the BARD Moodle training site.
There will be two options for the problem to be worked on.
Participants can propose a problem prior to the workshop, by emailing us at bardhelp@gmail.com by Friday 4 December. Please provide a short (2-3 paragraph) description of the problem, as well as the reason why you'd like to work on it, and (if relevant) which other attendees (e.g. from your organisation) that would also like to work collaboratively on this with you. We'll decide ahead of time if we think the problem (or at least a subset of this) is suitable for working on during the workshop.
A simple medical problem that we've used in BARD training before and that we know is suitable for learning how to use BARD.
Notes:
* This workshop assumes basic knowledge of Bayesian networks (for example, as given in the ABNMS Introduction to BNs workshop)
* The presentation part of the webinar will be recorded and made available on request afterwards