A new method is proposed for exploiting causal independencies in exact Bayesian network inference.

Existing data sets of cases can significantly reduce the knowledge engineering effort required to parameterize Bayesian networks. Unfortunately, when a data set is small, many conditioning cases are represented by too few or no data records and they do not offer sufficient basis for learning conditional probability distributions. We propose a method that uses Noisy-OR gates to reduce the data requirements in learning conditional probabilities. We test our method on HEPAR II, a model for diagnosis of liver disorders, whose parameters are extracted from a real, small set of patient records. Diagnostic accuracy of the multiple-disorder model enhanced with the Noisy-OR parameters was around 6% better than the accuracy of the plain multiple-disorder model and 10% better than the singledisorder diagnosis model.

Probabilistic networks are now fairly well established as practical representations of knowledge for reasoning under uncertainty, as demonstrated by an increasing number of successful applications in such domains as (medical) diagnosis and prognosis, planning, vision,

In designing a Bayesian network for an actual problem, developers need to bridge the gap between the mathematical abstractions o#ered by the Bayesian-network formalism and the features of the problem to be modelled. Qualitative probabilistic networks (QPNs) have been put forward as qualitative analogues to Bayesian networks, and allow modelling interactions in terms of qualitative signs. They thus have the advantage that developers can abstract from the numerical detail, and therefore the gap may not be as wide as for their quantitative counterparts. A notion that has been suggested in the literature to facilitate Bayesian-network development is causal independence. It allows exploiting compact representations of probabilistic interactions among variables in a network. In the paper, we deploy both causal independence and QPNs in developing and analysing a collection of qualitative, causal interaction patterns, called QC patterns. These are endowed with a fixed qualitative semantics, and are intended to o#er developers a highlevel starting point when developing Bayesian networks.

In designing a Bayesian network for an actual problem, developers need to bridge the gap between the mathematical abstractions offered by the Bayesian-network formalism and the features of the problem to be modelled. A notion that has been suggested in the literature to facilitate Bayesian-network development is causal independence. It allows exploiting compact representations of probabilistic interactions among variables in a network. However, only very few types of causal independence are in use today, as only the most obvious ones are really understood. We believe that qualitative probabilistic networks (QPNs) may be useful in helping understand causal independence. Originally, QPNs have been put forward as qualitative analogues to Bayesian networks. In this paper, we deploy QPNs in developing and analysing a collection of qualitative, causal interaction patterns, called QC patterns. These are endowed with a fixed qualitative semantics, and are intended to offer developers a high-level starting point when developing Bayesian networks.

This chapter reviews techniques for learning causal relationships from data, in application to the problem of feature selection. Most feature selection methods do not attempt to uncover causal relationships between feature and target and focus instead on making best predictions. We examine situations in which the knowledge of causal relationships benefits feature selection. Such benefits may include: explaining relevance in terms of causal mechanisms, distinguishing between actual features and experimental artifacts, predicting the consequences of actions performed by external agents, and making predictions in non-stationary environments. Conversely, we highlight the benefits that causal discovery may draw from recent developments in feature selection theory and algorithms.

The certainty-factor model was one of the most popular models for the representation and manipulation of uncertain knowledge in the early rule-based expert systems of the 1980s. After the model was criticised by researchers in artificial intelligence and statistics as being adhoc in nature, researchers and developers have stopped looking at the model. Nowadays, it is often stated that the model is merely interesting from a historical point of view. Its place has been taken over by more expressive formalisms for the representation and manipulation of uncertain knowledge, in particular, by the formalism of Bayesian belief networks. In this paper, it is shown that this view underestimates the importance of the principles underlying the certainty-factor model. In particular, it is shown that certainty-factor-like structures occur frequently in practical Bayesian network models as causal independence assumptions. In fact, the noisy-OR and noisy-AND models, two probabilistic models frequently employed, appear to be reinventions of combination functions previously introduced as part of the certaintyfactor model. This insight may lead to a reappraisal of the certainty-factor model. 2001 Elsevier Science B.V. All fights reserved.