Ann E. Nicholson

Parameterisation and evaluation of a Bayesian network for use in an ecological risk assessment

Catchment managers face considerable challenges in managing ecological assets. This task is made difficult by the variable and complex nature of ecological assets, and the considerable uncertainty involved in quantifying how various threats and hazards impact upon them. Bayesian approaches have the potential to address the modelling needs of environmental management. However, to date many Bayesian networks (Bn) developed for environmental management have been parameterised using knowledge elicitation only. Not only are these models highly qualitative, but the time and effort involved in elicitation of a complex Bn can often be overwhelming. Unfortunately in environmental applications, data alone are often too limited for parameterising a Bn. Consequently, there is growing interest in how to parameterise Bns using both data and elicited information. At present, there is little formal guidance on how to combine what can be learned from the data with what can be elicited. In a previous publication we proposed a detailed methodology for this process, focussing on parameterising and evaluating a Bn. In this paper, we further develop this methodology using a risk assessment case study, with the focus being on native fish communities in the Goulburn Catchment (Victoria, Australia).

Predicting users' requests on the WWW

We describe several Markov models derived from the behaviour patterns of many users, which predict which documents a user is likely to request next. We then present comparative results of the predictive accuracy of the different models, and, based on these results, build hybrid models which combine the individual models in different ways. These hybrid models generally have a greater predictive accuracy than the individual models. The best models will be incorporated in a system for pre-sending WWW documents.

Towards a Bayesian Model for Keyhole Plan Recognition in Large Domains

We present an approach to keyhole plan recognition which uses a Dynamic Belief Network to represent features of the domain that are needed to identify users’ plans and goals. The structure of this network was determined from analysis of the domain. The conditional probability distributions are learned during a training phase, which dynamically builds these probabilities from observations of user behaviour. This approach allows the use of incomplete, sparse and noisy data during both training and testing. We present experimental results of the application of our system to a Multi-User Dungeon adventure game with thousands of possible actions and positions. These results show a high degree of predictive accuracy and indicate that this approach will work in other domains with similar features.

Using Bayesian belief networks for change impact analysis in architecture design

Research into design rationale in the past has focused on argumentation-based design deliberations. These approaches cannot be used to support change impact analysis effectively because the dependency between design elements and decisions are not well represented and cannot be quantified. Without such knowledge, designers and architects cannot easily assess how changing requirements and design decisions may affect the system. In this article, we introduce the Architecture Rationale and Element Linkage (AREL) model to represent the causal relationships between architecture design elements and decisions. We apply Bayesian Belief Networks (BBN) to AREL, to capture the probabilistic causal relationships between design elements and decisions. We employ three different BBN-based reasoning methods to analyse design change impact: predictive reasoning, diagnostic reasoning and combined reasoning. We illustrate the application of the BBN modelling and change impact analysis methods by using a partial design of a real-world cheque image processing system. To support its implementation, we have developed a practical, integrated tool set for the architects to use.

Parameterising bayesian networks

Most documented Bayesian network (BN) applications have been built through knowledge elicitation from domain experts (DEs) The difficulties involved have led to growing interest in machine learning of BNs from data There is a further need for combining what can be learned from the data with what can be elicited from DEs In this paper, we propose a detailed methodology for this combination, specifically for the parameters of a BN.

Incorporating expert knowledge when learning Bayesian network structure: A medical case study

OBJECTIVES Bayesian networks (BNs) are rapidly becoming a leading technology in applied Artificial Intelligence, with many applications in medicine. Both automated learning of BNs and expert elicitation have been used to build these networks, but the potentially more useful combination of these two methods remains underexplored. In this paper we examine a number of approaches to their combination when learning structure and present new techniques for assessing their results. METHODS AND MATERIALS Using public-domain medical data, we run an automated causal discovery system, CaMML, which allows the incorporation of multiple kinds of prior expert knowledge into its search, to test and compare unbiased discovery with discovery biased with different kinds of expert opinion. We use adjacency matrices enhanced with numerical and colour labels to assist with the interpretation of the results. We present an algorithm for generating a single BN from a set of learned BNs that incorporates user preferences regarding complexity vs completeness. These techniques are presented as part of the first detailed workflow for hybrid structure learning within the broader knowledge engineering process. RESULTS The detailed knowledge engineering workflow is shown to be useful for structuring a complex iterative BN development process. The adjacency matrices make it clear that for our medical case study using the IOWA dataset, the simplest kind of prior information (partially sorting variables into tiers) was more effective in aiding model discovery than either using no prior information or using more sophisticated and detailed expert priors. The method for generating a single BN captures relationships that would be overlooked by other approaches in the literature. CONCLUSION Hybrid causal learning of BNs is an important emerging technology. We present methods for incorporating it into the knowledge engineering process, including visualisation and analysis of the learned networks.

Bayesian Networks for Clinical Decision Support in Lung Cancer Care

Survival prediction and treatment selection in lung cancer care are characterised by high levels of uncertainty. Bayesian Networks (BNs), which naturally reason with uncertain domain knowledge, can be applied to aid lung cancer experts by providing personalised survival estimates and treatment selection recommendations. Based on the English Lung Cancer Database (LUCADA), we evaluate the feasibility of BNs for these two tasks, while comparing the performances of various causal discovery approaches to uncover the most feasible network structure from expert knowledge and data. We show first that the BN structure elicited from clinicians achieves a disappointing area under the ROC curve of 0.75 (± 0.03), whereas a structure learned by the CAMML hybrid causal discovery algorithm, which adheres with the temporal restrictions, achieves 0.81 (± 0.03). Second, our causal intervention results reveal that BN treatment recommendations, based on prescribing the treatment plan that maximises survival, can only predict the recorded treatment plan 29% of the time. However, this percentage rises to 76% when partial matches are included.

Anomaly detection in vessel tracks using Bayesian networks

Abstract In recent years electronic tracking has provided voluminous data on vessel movements, leading researchers to try various data mining techniques to find patterns and, especially, deviations from patterns, i.e., for anomaly detection. Here we describe anomaly detection with data mined Bayesian Networks, learning them from real world Automated Identification System (AIS) data, and from supplementary data, producing both dynamic and static Bayesian network models. We find that the learned networks are quite easy to examine and verify despite incorporating a large number of variables. We also demonstrate that combining dynamic and static modelling approaches improves the coverage of the overall model and thereby anomaly detection performance.

Seabreeze Prediction Using Bayesian Networks

In this paper we examine the use of Bayesian networks (BNs) for improving weather prediction, applying them to the problem of predicting sea breezes. We compare a pre-existing Bureau of Meteorology rule-based system with an elicited BN and others learned by two data mining programs, TETRAD II [Spirtes et al., 1993] and Causal MML [Wallace and Korb, 1999]. These Bayesian nets are shown to significantly outperform the rule-based system in predictive accuracy.

Faster and more accurate classification of time series by exploiting a novel dynamic time warping averaging algorithm

A concerted research effort over the past two decades has heralded significant improvements in both the efficiency and effectiveness of time series classification. The consensus that has emerged in the community is that the best solution is a surprisingly simple one. In virtually all domains, the most accurate classifier is the nearest neighbor algorithm with dynamic time warping as the distance measure. The time complexity of dynamic time warping means that successful deployments on resource-constrained devices remain elusive. Moreover, the recent explosion of interest in wearable computing devices, which typically have limited computational resources, has greatly increased the need for very efficient classification algorithms. A classic technique to obtain the benefits of the nearest neighbor algorithm, without inheriting its undesirable time and space complexity, is to use the nearest centroid algorithm. Unfortunately, the unique properties of (most) time series data mean that the centroid typically does not resemble any of the instances, an unintuitive and underappreciated fact. In this paper we demonstrate that we can exploit a recent result by Petitjean et al. to allow meaningful averaging of “warped” time series, which then allows us to create super-efficient nearest “centroid” classifiers that are at least as accurate as their more computationally challenged nearest neighbor relatives. We demonstrate empirically the utility of our approach by comparing it to all the appropriate strawmen algorithms on the ubiquitous UCR Benchmarks and with a case study in supporting insect classification on resource-constrained sensors.