Literature

A brief selection of peer-reviewed articles relevant to Bayesian Network modelling containing either key methodological advances or else examples of application.


General note

One general point of note is that typical BN models involving binary nodes, arguably the most commonly used type of BN, use a contingency table rather than additive parameter formulation. This facilities mathematical elegance and means that key metrics like model goodness of fit and marginal posterior parameters can be estimated analytically (e.g. from a formula) rather than numerically (an approximation). The downside being that this parameterisation is likely far from parsimonious, and the interpretation of the model parameters is less clear than in more usual GLM type models (which are common across all areas of science). This is, while practically important, a fairly low level technical distinction as the key aspect of BN modelling is that this is a form of graphical modelling – that is a model of the joint probability distribution of the data. It is this joint – multidimensional – aspect which makes this methodology so attractive for analyses of complex data and what discriminates it from the more standard regression techniques, e.g. glm’s, glmm’s etc, which are only one dimensional in that the covariates are all assumed independent. The latter is entirely reasonable in a classical experimental design scenario, but completely unrealistic for many observational studies in medicine, ecology and biology.


Key technical/theoretical articles

  • Author: Heckerman, D. and Geiger, D. and Chickering, D. M.
    Title: Learning Bayesian Networks – The Combination of Knowledge And Statistical-Data
    Journal: Machine Learning
    Year: 1995
    Volume: 20
    Pages: 197-243
    link to article pdf
    comments: Arguably the founding article in Bayesian Network modelling.
  • Author: Friedman, N. and Koller, D.
    Title: Being Bayesian about network structure. A Bayesian approach to structure discovery in Bayesian networks
    Journal: Machine Learning
    Year: 2003
    Volume: 50
    pages: 95-125
    link to article pdf
    comments: This order-based approach was a big step forward in terms of dealing with larger problems (more nodes) and the difficulty of sampling across different DAGs.
  • Author: Friedman, N. and Goldszmidt, M. and Wyner, A.
    Title: Data analysis with Bayesian networks: A bootstrap approach
    Journal: Proc. Fifteenth Conference on Uncertainty in Artificial Intelligence (UAI’99) (pp.206-215). San Francisco: Morgan Kaufmann
    Year: 1999
    link to article pdf
    comments: Introduces parametric and non-parametric bootstrapping for model selection in BN analyses.
  • Author: Koivisto, M. and Sood, K.
    Title: Exact Bayesian structure discovery in Bayesian networks
    Journal: Journal of Machine Learning Research
    Year: 2004
    Volume: 5
    Pages: 549-573
    link to article pdf
    comments: Extends the elegant order based approach from Friedman and Koller into an exact setting – exact posterior probabilities and the globally optimal DAG can be identified – provided the dimension (number of variables) is not too large.

Application/case study articles

  • Author: Lewis, F. I. and McCormick, B. J. J.
    Title: Revealing the Complexity of Health Determinants in Resource-poor Settings
    Journal: American Journal of Epidemiology
    Year: 2012
    Volume: 176(11)
    Pages: [DOI: 10.1093/aje/kws183]
    link to article pdf
    comments: Application of using additive Bayesian networks to explore risk factors for diarrhea in children
  • Author: Jansen, R., (Yu, H. Y., Greenbaum, D., Kluger, Y., Krogan, N. J., Chung, S. B., Emili, A., Snyder, M., Greenblatt, J. F., Gerstein, M.
    Title: A Bayesian networks approach for predicting protein-protein interactions from genomic data
    Journal: Science
    Year: 2003
    Volume: 302(5644)
    Pages: 449-453
    link to article pdf
    comments: Application of using additive Bayesian networks to predict results in noisy data
  • Author: Poon, A. F. Y., Lewis, F. I., Kosakovsky Pond, S. L., Frost, S. D. W.
    Title: An Evolutionary-Network Model Reveals Stratified Interactions in the V3 Loop of the HIV-1 Envelope
    Journal: PLoS Computational Biology
    Year: 2007
    Volume: 3(11)
    Pages: [doi:10.1371/journal.pcbi.0030231]
    link to article pdf
    comments: Application of using Bayesian networks in HIV
  • Author: Lewis, F. I., BrĂ¼lisauer, F. and Gunn, G.J.
    Title: Structure discovery in Bayesian networks: An analytical tool for analysing complex animal health data
    Journal: Preventive Veterinary Medicine
    Year: 2011
    Volume: 100
    Pages:109-115
    link to article pdf
    comments: Application of using Bayesian networks in veterinary epidemiology
  • Author: Hartnack, S., Springer, S., Pittavino, M. and Grimm, H.
    Title: Attitudes of Austrian veterinarians towards euthanasia in small animal practice: impacts of age and gender on views on euthanasia
    Journal: BMC Veterinary Research
    Year: 2016
    Volume: 12
    Pages:1-14
    link to article pdf
    comments: Application of using additive Bayesian networks in veterinary epidemiology, to untangle attitudes of Austrian veterinarians

Recent Posts