this paper, we propose an automated process for constructing the combined dependency structure of a ########## probabilistic network. Each domain expert supplies any known conditional independency information and not necessarily an explicit dependency structure. Our method determines a succinct representation of all the supplied independency information called a ####### #####. This process involves detecting all ############ information and removing all ######### information. A ###### dependency structure of the multiagent probabilistic network can be constructed directly from this minimal cover. The main result of this paper is that the constructed dependency structure is a ########### of the minimal cover. That is, every probabilistic conditional independency logically implied by the minimal cover can be inferred from the dependency structure and every probabilistic conditional independency inferred from the dependency structure is logically implied by the minimal cover

In this paper, we consider the problem of performing inference on Bayesian networks which exhibit a substantial degree of determinism.

Data dependencies are used in database schema design to enforce the correctness of a database as well as to reduce redundant data. These dependencies are usually determined from the semantics of the attributes and are then enforced upon the relations. This paper describes a bottom-up procedure for discovering multivalued dependencies (MVDs) in observed data without knowing `a priori the relationships amongst the attributes. The proposed algorithm is an application of the technique we designed for learning conditional independencies in probabilistic reasoning. A prototype system for automated database schema design has been implemented. Experiments were carried out to demonstrate both the effectiveness and efficiency of our method. 1

Data dependencies have been extensively studied in relational databases as they play a key role in the normalization process. On the other hand, probabilistic reasoning systems would not be practical without the notion of probabilistic conditional independence. In this paper, we present a detailed comparison of these two types of (in)dependencies. While past research has shown that multivalued dependency is a necessary but not sufficient condition for conditional independence, here we show in particular that functional dependency is a sufficient but not necessary condition for conditional independence.

In this paper, a model is proposed for multi-agent probabilistic reasoning in a distributed environment. Unlike other methods, this model is capable of processing input in a truly asynchronous fashion. Asynchronous control protocols and a method for processing evidence are developed to ensure global consistency at all times. The proposed system then extends beyond an interpretive system since the now well-defined concept of a distributed request can be introduced. Techniques are also suggested to reduce data transmission in answering this type of request.

In this paper, we introduce aprobabilistic relational data model as the basis for developing multi-agent probabilistic reasoning systems. Since our model subsumes the traditional relational data model, it immediately follows that we can take full advantage of the existing distributed and concurrency control techniques to address the undesirable characteristics exhibited by current multi-agent probabilistic reasoning systems. Thereby, our probabilistic relational data model has important theoretical and practical rami cations. One uni ed model allows the cross-fertilization of techniques, and serves as a basis for implementing one system for both of these similar domains.

In our earlier paper, we introduced the notion of granular probabilistic networks. We use the term granular to mean the ability to coarsen and refine parts of a joint probability distribution. In practice, however, the joint distribution is usually represented as a product of marginal distributions. In this paper, we show that the coarsening operation can be applied locally to the marginal distributions with the same effect as if applied to the joint distribution. This is an important result as otherwise the joint distribution would have to be computed before performing the coarsening operation.

Based on the elegant theory of relational databases, the present investigation establishes a unified model for both relational databases and Bayesian networks. This is in contradiction to the argument that relational databases and Bayesian networks are different, where it was shown that the implication problem does not coincide for embedded multivalued dependency (EMVD) and probabilistic conditional independence (CI). The main result of this thesis, however, is that the implication problem coincides on the solvable subclasses of EMVD and CI, but differs on the unsolvable general classes of EMVD and CI. This means that there is no practical difference between relational databases and Bayesian networks, since only the solvable subclasses are useful in the design of both of these knowledge systems.

Markov networks utilize nonembedded probabilistic conditional independencies in order to provide an economical representation of a joint distribution in uncertainty management. In this paper we study several properties of nonembedded conditional independencies and show that they are in fact equivalent. The results presented here not only show the useful characteristics of an important subclass of probabilistic conditional independencies, but further demonstrate the relationship between relational theory and probabilistic reasoning. 1

In this paper, we study the problem of triangulation of Bayesian networks from a relational database perspective. We show that the problem of triangulating a Bayesian network is equivalent to the problem of identifying a maximal subset of conflict free conditional independencies.