Bayesian networks (BNs) are a widely used probabilistic modelling tool for reasoning under uncertainty, though scaling them up for complex real-world problems can be challenging. Object-Oriented Bayesian Networks (OOBNs) have been proposed to address this challenge, providing modellers with the ability to define hierarchies of classes and use these classes to construct models with a compositional and hierarchical structure, enabling reuse and supporting maintenance. The object-oriented concept of inheritance supports reuse of existing components, but comes with the challenge of building, and then maintaining, an efficient hierarchy of classes. This paper proposes a supergraph based method that constructs class inheritance hierarchies from a set of OOBN classes; this can be used either to form an initial inheritance hierarchy, or to reform an existing hierarchy into a more efficient one. We also present heuristics to convert a BN to an OOBN class, measures to evaluate a constructed hierarchy and empirical analyses of the proposed approach on synthetic hierarchies and on a real-world OOBN project; results show the algorithm works well in practice.

The construction of Bayesian Networks (BNs) to model large-scale real-life problems is challenging. One approach to scaling up is Object Oriented Bayesian Networks (OOBNs). These provide modellers with the ability to define classes and construct models with a compositional and hierarchical structure, enabling reuse and supporting maintenance. In the OO programming paradigm, a key concept is inheritance, the ability to derive attributes and behavior from pre-existing classes, which enables an even higher level of reusability and scalability. However, inheritance in OOBNs has yet to be fully defined and implemented. Here we present iOOBN, a tool which provides fully defined inheritance for OOBNs. We provide guidance on modelling in iOOBN, describe our prototype implementation with an existing BN software tool, Hugin, and demonstrate its applicability and usefulness via a case study of re-engineering an existing large complex dynamic OOBN.

Recent years have seen significant progress in improving both the efficiency and effectiveness of time series classification. However, because the best solution is typically the Nearest Neighbor algorithm with the relatively expensive Dynamic Time Warping as the distance measure, successful deployments on resource constrained devices remain elusive. Moreover, the recent explosion of interest in wearable devices, which typically have limited computational resources, has created a growing need for very efficient classification algorithms. A commonly used technique to glean the benefits of the Nearest Neighbor algorithm, without inheriting its undesirable time complexity, is to use the Nearest Centroid algorithm. However, because of the unique properties of (most) time series data, the centroid typically does not resemble any of the instances, an unintuitive and underappreciated fact. In this work we show that we can exploit a recent result to allow meaningful averaging of 'warped' times series, and that this result allows us to create ultra-efficient Nearest 'Centroid' classifiers that are at least as accurate as their more lethargic Nearest Neighbor cousins.

Association discovery is a fundamental data mining task. The primary statistical approach to association discovery between variables is log-linear analysis. Classical approaches to log-linear analysis do not scale beyond about ten variables. We develop an efficient approach to log-linear analysis that scales to hundreds of variables by melding the classical statistical machinery of log-linear analysis with advanced data mining techniques from association discovery and graphical modeling.

In a deterministic world, a planning agent can be certain of the consequences of its planned sequence of actions. Not so, however, in dynamic, stochastic domains where Markov decision processes are commonly used. Unfortunately these suffer from the `curse of dimensionality': if the state space is a Cartesian product of many small sets (`dimensions'), planning is exponential in the number of those dimensions. Our new technique exploits the intuitive strategy of selectively ignoring various dimensions in different parts of the state space. The resulting non-uniformity has strong implications, since the approximation is no longer Markovian, requiring the use of a modified planner. We also use a spatial and temporal proximity measure, which responds to continued planning as well as movement of the agent through the state space, to dynamically adapt the abstraction as planning progresses. We present qualitative and quantitative results across a range of experimental domains showing that an agent exploiting this novel approximation method successfully finds solutions to the planning problem using much less than the full state space. We assess and analyse the features of domains which our method can exploit.

A Bayesian Network (BN) consists of a qualitative part representing the structural assumptions of the domain and a quantitative part, the parameters. To date, knowledge engineering support has focused on parameter elicitation, with little support for designing the graphical structure. Poor design choices in BN construction can impact the network's performance, network maintenance, and the explanatory power of the output. We present a tool to help domain experts examine BN structure independently of the parameters. Our qualitative evaluation of the tool shows that it can help in identifying possible structural modeling errors and, hence, improve the quality of BN models. © 2006 Wiley Periodicals, Inc. Int J Int Syst 21: 1127–1150, 2006.

Research into design rationale in the past has focused on the representation of reasons and has omitted the connections between design rationales and design artefacts. Without such connections, designers and architects cannot easily assess how changing requirements or designs may affect the system. In this paper, we introduce a model called Architecture Rationale and Element Linkage (AREL) to represent the causal relationship between architecture elements and decisions. We further model AREL as a Bayesian Belief Network (BBN) to capture the probabilistic relationships between architecture elements and decisions in an architecture design model. Such probabilistic modelling enables architects to quantitatively predict and diagnose impact of change when part of the requirements or designs are changing. Using a partial design of a cheque image processing system, we illustrate how AREL is used to represent the decision model and how BBN is used to predict and diagnose change in the architecture design. We use a UML tool to capture the AREL model and a BBN tool to compute the probabilities of change impact.

We explore the problem of how to perform complex strategic planning in an uncertain domain. Although card games such as bridge and poker provide interesting domains for exploring uncertainty, they tend to be strategically simple given perfect information. We present and discuss a strategically complex, easily controlled, imperfect information domain-a variant of chess-which allows us to explore this difficult problem. We also describe a modular hybrid architecture that uses game theory, information theory and utility theory to play this new game. We discuss preliminary results from our architecture and consider possible future directions for the project.

We describe the development of a monitoring system which uses sensor observation data about discrete events to construct dynamically a probabilistic model of the world. This model is a Bayesian network incorporating temporal aspects, which we call a dynamic belief network; it is used to reason under uncertainty about both the causes and consequences of the events being monitored. The basic dynamic construction of the network is data-driven. However the model construction process combines sensor data about events with externally provided information about agents' behavior, and knowledge already contained within the model, to control the size and complexity of the network. This means that both the network structure within a time interval, and the amount of history and detail maintained, can vary over time. We illustrate the system with the example domain of monitoring robot vehicles and people in a restricted dynamic environment using light-beam sensor data. In addition to presenting a generic network structure for monitoring domains, we describe the use of more complex network structures which address two specific monitoring problems, sensor validation and the data association problem.>