Learning Bayesian networks: The combination of knowledge and statistical data

@article{Heckerman1995LearningBN,
  title={Learning Bayesian networks: The combination of knowledge and statistical data},
  author={David Heckerman and Dan Geiger and David Maxwell Chickering},
  journal={Machine Learning},
  year={1995},
  volume={20},
  pages={197-243}
}
We describe a Bayesian approach for learning Bayesian networks from a combination of prior knowledge and statistical data. First and foremost, we develop a methodology for assessing informative priors needed for learning. Our approach is derived from a set of assumptions made previously as well as the assumption oflikelihood equivalence, which says that data should not help to discriminate network structures that represent the same assertions of conditional independence. We show that likelihood… Expand

Topics from this paper

Data mining tasks and methods: Probabilistic and casual networks: methodology for probabilistic networks
This article provides an overview of how to handle uncertainty about which Bayesian network to use for calculating the effect of an ideal manipulation or a classification. The Bayesian approach toExpand
Learning Bayesian Networks: The Combination of Knowledge and Statistical Data
TLDR
A methodology for assessing informative priors needed for learning Bayesian networks from a combination of prior knowledge and statistical data is developed and how to compute the relative posterior probabilities of network structures given data is shown. Expand
A-2015-2 Approximation Strategies for Structure Learning in Bayesian Networks
Bayesian networks are probabilistic graphical models, which can compactly represent complex probabilistic dependencies between a set of variables. Once learned from data or constructed by some otherExpand
LEARNING LIKELIHOOD-EQUIVALENCE BAYESIAN NETWORKS USING AN EMPIRICAL BAYESIAN APPROACH
Many studies on learning Bayesian networks have used the Dirichlet prior score metric (DPSM). Although they assume different optimum hyper-parameter values for DPSM, few studies have focused onExpand
Who learns better Bayesian network structures: Accuracy and speed of structure learning algorithms
TLDR
It is found using both simulated and real-world complex data that constraint- based algorithms are often less accurate than score-based algorithms, but are seldom faster (even at large sample sizes); and that hybrid algorithms are neither faster nor more accurate than constraint-based algorithm. Expand
Using Weak Prior Information on Structures to Learn Bayesian Networks
TLDR
An elicitation procedure for DAGs is proposed which exploits prior knowledge on network topology, and that is suited to large Bayesian Networks, and a new quasi-Bayesian score function is developed, the P-metric, to perform structural learning following a score-and-search approach. Expand
Adaptive Probabilistic Networks with Hidden Variables
TLDR
This paper presents a gradient-based algorithm and shows that the gradient can be computed locally, using information that is available as a byproduct of standard inference algorithms for probabilistic networks. Expand
Encoding structural prior information to learn large Bayesian Networks
Most of the approaches developed in the literature to elicit the a-priori distribution on Directed Acyclic Graphs (DAGs) require a full specification of graphs. Nevertheless, expert’s prior knowledgeExpand
Learning bayesian networks for solving real-world problems
TLDR
This thesis presents a principled method, based on the EM algorithm, for learning both Bayesian network structure and probabilities from incomplete data, and evaluates its performance on several datasets with di erent amounts of missing data and assumptions about the missing data mechanisms. Expand
Bayesian data mining and knowledge discovery
One of the major problems faced by data-mining technologies is how to deal with uncertainty. The prime characteristic of Bayesian methods is their explicit use of probability for quantifyingExpand
...
1
2
3
4
5
...

References

SHOWING 1-10 OF 44 REFERENCES
Learning Bayesian Networks: Search Methods and Experimental Results
TLDR
A metric for computing the relative posterior probability of a network structure given data developed by Heckerman et al. (1994a,b,c) has a property useful for inferring causation from data and is described. Expand
Learning Gaussian Networks
TLDR
This work extends traditional statistical approaches for identifying vanishing regression coefficients in that it identifies two important assumptions, called event equivalence and parameter modularity, that when combined allow the construction of prior distributions for multivariate normal parameters from a single prior Bayesian network specified by a user. Expand
A Bayesian Approach to Learning Causal Networks
TLDR
This paper introduces two sufficient assumptions, called mechanism independence and component independence, and shows that these new assumptions, when combined with parameter independence, parameter modularity, and likelihood equivalence, allow methods for learning acausal networks to learn causal networks. Expand
Learning Bayesian Networks: A Unification for Discrete and Gaussian Domains
TLDR
A general Bayesian scoring metric is derived, appropriate for both discrete and Gaussian domains, from well-known statistical facts about the Dirichlet and normal--Wishart distributions. Expand
An Algorithm for the Construction of Bayesian Network Structures from Data
TLDR
An algorithm is presented that integrates two approaches to the construction of Bayesian belief network structures from data - CI tests are used to generate an ordering on the nodes from the database which is then used to recover the underlying Bayesian network structure using a non CI based method. Expand
Using Causal Information and Local Measures to Learn Bayesian Networks
TLDR
A new local way of computing the description length is presented, which allows for significant improvements in the search algorithm and opens the door for local refinement of an existent network. Expand
Bayesian analysis in expert systems
TLDR
Using a real, moderately complex, medical example, it is illustrated how qualitative and quantitative knowledge can be represented within a directed graphical model, generally known as a belief network in this context. Expand
The Assessment of Prior Distributions in Bayesian Analysis
Abstract In the Bayesian framework, quantified judgments about uncertainty are an indispensable input to methods of statistical inference and decision. Ultimately, all components of the formalExpand
Model Selection and Accounting for Model Uncertainty in Graphical Models Using Occam's Window
Abstract We consider the problem of model selection and accounting for model uncertainty in high-dimensional contingency tables, motivated by expert system applications. The approach most usedExpand
A Construction of Bayesian Networks from Databases Based on an MDL Principle
  • J. Suzuki
  • Computer Science, Mathematics
  • UAI
  • 1993
TLDR
The extended version of the algorithm of Chow and Liu in that the learning algorithm selects the model in the range where the dependencies among the attributes are represented by some general plural number of trees. Expand
...
1
2
3
4
5
...