This literature review discusses different methods under the general rubric of learning Bayesian networks from data, and includes some overlapping work on more general probabilistic networks. Connections are drawn between the statistical, neural network, and uncertainty communities, and between the different methodological communities, such as Bayesian, description length, and classical statistics. Basic concepts for learning and Bayesian networks are introduced and methods are then reviewed. Methods are discussed for learning parameters of a probabilistic network, for learning the structure, and for learning hidden variables. The presentation avoids formal definitions and theorems, as these are plentiful in the literature, and instead illustrates key concepts with simplified examples. Keywords--- Bayesian networks, graphical models, hidden variables, learning, learning structure, probabilistic networks, knowledge discovery. I. Introduction Probabilistic networks or probabilistic gra...

Algorithms for learning Bayesian networks from data have two components: a scoring metric and a search procedure. The scoring metric computes a score reflecting the goodness-of-fit of the structure to the data. The search procedure tries to identify network structures with high scores. Heckerman et al. (1994) introduced a Bayesian metric, called the BDe metric, that computes the relative posterior probability of a network structure given data. They show that the metric has a property desireable for inferring causal structure from data. In this paper, we show that the problem of deciding whether there is a Bayesian network---among those where each node has at most k parents---that has a relative posterior probability greater than a given constant is NP-complete, when the BDe metric is used. 1 Introduction Recently, many researchers have begun to investigate methods for learning Bayesian networks, including Bayesian methods [Cooper and Herskovits, 1991, Buntine, 1991, York 1992, Spiegel...

This paper provides algorithms that use an information-theoretic analysis to learn Bayesian network structures from data. Based on our three-phase learning framework, we develop efficient algorithms that can effectively learn Bayesian networks, requiring only polynomial numbers of conditional independence (CI) tests in typical cases. We provide precise conditions that specify when these algorithms are guaranteed to be correct as well as empirical evidence (from real world applications and simulation tests) that demonstrates that these systems work efficiently and reliably in practice.

In this work we introduce a methodology based on Genetic Algorithms for the automatic induction of Bayesian Networks from a file containing cases and variables related to the problem. The methodology is applied to the problem of predicting survival of people after one, three and five years of being diagnosed as having malignant skin melanoma. The accuracy of the obtained model, measured in terms of the percentage of well-classified subjects, is compared to that obtained by the called Naive-Bayes. In both cases, the estimation of the model accuracy is obtained from the 10-fold cross-validation method. 1. Introduction Expert systems, one of the most developed areas in the field of Artificial Intelligence, are computer programs designed to help or replace humans beings in tasks in which the human experience and human knowledge are scarce and unreliable. Although, there are domains in which the tasks can be specifed by logic rules, other domains are characterized by an uncertainty inherent...

In this paper we perform a comparison among FSS-EBNA, a randomized, populationbased and evolutionary algorithm, and two genetic and other two sequential search approaches in the well known Feature Subset Selection (FSS) problem. In FSS-EBNA, the FSS problem, stated as a search problem, uses the EBNA (Estimation of Bayesian Network Algorithm) search engine, an algorithm within the EDA (Estimation of Distribution Algorithm) approach. The EDA paradigm is born from the roots of the GA community in order to explicitly discover the relationships among the features of the problem and not disrupt them by genetic recombination operators. The EDA paradigm avoids the use of recombination operators and it guarantees the evolution of the population of solutions and the discovery of these relationships by the factorization of the probability distribution of best individuals in each generation of the search. In EBNA, this factorization is carried out by a Bayesian network induced by a chea...

Estimation of Distribution Algorithms (EDA) constitute an example of stochastics heuristics based on populations of individuals every of which encode the possible solutions to the optimization problem. These populations of individuals evolve in succesive generations as the search progresses -- organized in the same way as most evolutionary computation heuristics. In opposition to most evolutionary computation paradigms which consider the crossing and mutation operators as essential tools to generate new populations, EDA replaces those operators by the estimation and simulation of the joint probability distribution of the selected individuals. In this work, after making a review of the different approaches based on EDA for problems of combinatorial optimization as well as for problems of optimization in continuous domains, we propose new approaches based on the theory of probabilistic graphical models to solve problems in both domains. More precisely, we propose to adapt algorit...

We consider a Bayesian method for learning the Bayesian network structure from complete data. Recently, Koivisto and Sood (2004) presented an algorithm that for any single edge computes its marginal posterior probability in O(n2 n) time, where n is the number of attributes; the number of parents per attribute is bounded by a constant. In this paper we show that the posterior probabilities for all the n(n−1) potential edges can be computed in O(n2 n) total time. This result is achieved by a forward–backward technique and fast Möbius transform algorithms, which are of independent interest. The resulting speedup by a factor of about n 2 allows us to experimentally study the statistical power of learning moderate-size networks. We report results from a simulation study that covers data sets with 20 to 10,000 records over 5 to 25 discrete attributes. 1

This paper presents an efficient algorithm for constructing Bayesian belief networks from databases. The algorithm takes a database and an attributes ordering (i.e., the causal attributes of an attribute should appear earlier in the order) as input and constructs a belief network structure as output. The construction process is based on the computation of mutual information of attribute pairs. Given a data set which is large enough and has a DAGIsomorphic probability distribution, this algorithm guarantees that the perfect map [1] of the underlying dependency model is generated, and at the same time, enjoys the time complexity of O N ( ) on conditional independence (CI) tests. To evaluate this algorithm, we present the experimental results on three versions of the well-known ALARM network database, which has 37 attributes and 10,000 records. The correctness proof and the analysis of computational complexity are also presented. We also discuss the features of our work and relate it to previous works.

This paper addresses the problem of learning Bayesian network structures from data by using an information theoretic dependency analysis approach. Based on our three-phase construction mechanism, two efficient algorithms have been developed. One of our algorithms deals with a special case where the node ordering is given, the algorithm only require ) ( 2 N O CI tests and is correct given that the underlying model is DAG-Faithful [Spirtes et. al., 1996]. The other algorithm deals with the general case and requires ) ( 4 N O conditional independence (CI) tests. It is correct given that the underlying model is monotone DAG-Faithful (see Section 4.4). A system based on these algorithms has been developed and distributed through the Internet. The empirical results show that our approach is efficient and reliable. 1 Introduction The Bayesian network is a powerful knowledge representation and reasoning tool under conditions of uncertainty. A Bayesian network is a directed acyclic graph ...

We discuss two basic questions related to the synthesis of decision algorithms. The first question can be formulated as follows: what strategies can be used in order to discover the decision rules from experimental data? Answering this question, we propose to build these strategies on the basis of rough set methods and Boolean reasoning techniques. We present some applications of these methods for extracting decision rules from decision tables used to represent experimental data. The second question can be formulated as follows: what is a general framework for approximate reasoning in distributed systems? Answering this question, we assume that distributed systems are organized on rough mereological principles in order to assembly (construct) complex objects satisfying a given specification in a satisfactory degree. We discuss how this approach can be used for building the foundations for approximate reasoning. Our approach is based on rough mereology, the recently developed extensio...