Bayesian Intelligence - Bayesian Intelligence

Experts in Bayesian network and causal network solutions to real world modelling problems.

We	elicit and engineer Bayesian models,
	data mine Bayesian models,
	program and integrate Bayesian models,
	and train in the latest Bayesian network techniques.

More about our services

What are Bayesian networks?

Follow Bayesian Intelligence:

Understanding is about more than associations

At Bayesian Intelligence, we recognise you can go a long way with models based on associations. However, if you ever want to make a decision, associations will mislead, confuse, lie, and get you into all sorts of trouble.

To make a decision, you have to predict what will happen if you take action. And to predict what will happen, you have to be causal.

Bayesian networks let you do that. And they also let you predict, diagnose, understand, explain, work with missing data, use information from multiple places, mix expert information with data, generalise, quantify, discover...

Services

Our passion is modelling the world with Bayesian networks and causal models. We can help you define your problem, conceptualise, integrate expert knowledge, leverage big and small data, analyse risk and make models that provide key insights that support your most difficult and uncertain decisions.

Knowledge Engineering

We can help clients build Bayesian network models of their problem domains. We can work with domain experts to help articulate their understanding of a problem or system showing aspects of uncertainty. We have software tools available for simplifying and validating the elicitation of causal relationships, probabilistic dependencies and probabilities.

Causal Discovery

Causal discovery algorithms can automatically find the best Bayesian networks for explaining the relationships in data. These algorithms can be tuned using the results of knowledge engineering with experts to find useful solutions more rapidly. We also have experience with many other data mining methods, including predictive networks, classification trees and graphs and time series analysis.

Training

We offer training in all of the Bayesian network techniques we use, ranging from short courses to comprehensive introductory training, in-depth theoretical treatments to hands-on application training. We offer training on major Bayesian network software packages such as Netica and GeNIe, as well as programming with Bayesian networks for those more technically inclined.

Programming & Solutions

We can tailor tools to particular software environments or program stand-alone solutions. We can package any of these services together in whatever combination you need. We're also happy to provide ongoing maintenance, including support, enhancement and hosting of modelling solutions.

Enquire about our services

Partners, Collaborators and Clients

We have worked to develop solutions with and for partners from a range of government, academic and industry organisations.

Telethon Kids Institute Defence Science and Technology Department of Energy, Environment and Climate Action National Parks & Wildlife Service, New South Wales Health & Clinical Analytics, University of Sydney Ministry of Primary Industries, New Zealand Plant & Health Australia Plant & Food Research, New Zealand Scion Reseach, New Zealand Department of Agriculture, Fisheries and Forestry Australian Climate Service Bayesian Network Modelling Association Australian Securities & Insurance Commission Monash University Centre of Excellence for Biosecurity Risk Analysis New South Wales Rural Fire Service James Cook University Department of Transport and Planning, Victoria

Selected Projects

We've had the privilege of working on many interesting projects. Here is just a sample.

Supporting long term policies for climate change

As climate change worsens, predicting the consequences for human health becomes increasingly difficult, particularly as we try to project further into the future. The Australian Climate Service is exploring ways to deal with the deep uncertainty associated with such long term forecasts, and the difficulties of handling compounding risks — multiple problems that interact making outcomes worse than expected. We developed a flexible Bayesian approach making use of QGeNIe to perform the qualitative modelling required for long term forecasting. Through a sequence of expert workshops, we elicited a set of causal models that estimated the most significant risks for future human health, pointing the way to further analysis, improvements in policy and future adaptation strategies.

Actions for Biodiversity Conservation

Victoria's Department of Sustainability and Environment (DSE) uses the Actions for Biodiversity Conservation system for recording information on threatened species across the state. We created a software component (known as ABC-BN) that fully integrates into this database, allowing species managers to explicitly model the impacts of threats on species and the effectiveness of management actions used to mitigate those threats.

Restoring the Victorian Western Grasslands Reserves

The Victorian Government is reserving 15,000 hectares of land to the west and north-west of Werribee as a way of protecting endangered grasslands and offsetting the ecological costs of urban growth to the north and west of Melbourne. The DSE wishes to provide Parks Victoria with an adaptive management process for these grasslands, which will involve modelling, predicting, monitoring and revising the model in repsonse to discrepancies that arise in the review. DSE has chosen Bayesian networks as the modelling technology, and we are building the initial models to be used in this process, as well as providing training.

Understanding risk to improve market access and environmental outcomes

We are developing models for New Zealand's Scion research institute to minimise the risk of pests appearing in wood chips due for export. The aim is to sufficiently reduce the risk of attracting pests throughout the cutting, transportation and storage stages, such that costly and environmentally-unfriendly fumigation can be avoided prior to export.

Data mining anomalous vessel movements for defence security

We are creating software for the Defence Science and Technology Organisation based on our machine learning software CaMML and earlier research work to detect anomalous behaviour in track data sets of various kinds (such as ships and cars). This project involves mining large time series data sets in order to produce causal models that yield an understanding of normal behaviour, against which anomalies can be assessed. The emphasis in the current project will be to allow users of the software to create and test models and to provide detection in an online environment.

Prioritising power pole maintenance to improve grid stability

Western Power manages over 700,000 power poles in Perth and surrounding areas, many of which are nearing the end of their service lives. Western Power has recently embarked on a project aimed at improving the safety of these poles, while minimising the maintenance cost. The company settled on Bayesian networks as the best approach to model the health of the wood poles, given their strong risk management capabilities. We are providing Western Power with modelling advice and performance-focused BN programming solutions to assist in managing the risk of power pole failure.

Aggregating expert estimates for better political forecasting

The US government organisation IARPA established the ACE (Aggregative Contingent Estimation) program in order to develop and improve methods for eliciting conditional predictions from experts and also for aggregating those predictions into a single prediction. One of the teams participating in that program is a group called DAGGRE, based at George Mason University. Together with ACERA, we are assisting DAGGRE with the development of their combinatorial prediction market software, now based on an underlying BN back-end, along with their conditional and unconditional elicitation processes. We additionally created information gathering and elicitation software for ACERA's own elicitation and aggregated prediction experiments.