Subscribe

Issue No.12 - December (2008 vol.20)

pp: 1587-1600

Martin Stetter , SIEMENS AG., Munich

Rui Chang , Technical University Munich, Munich

DOI Bookmark: http://doi.ieeecomputersociety.org/10.1109/TKDE.2008.89

ABSTRACT

In this paper, we consider the problem of performing quantitative Bayesian inference and model averaging based on a set of qualitative statements about relationships. Statements are transformed into parameter constraints which are imposed onto a set of Bayesian networks. Recurrent relationship structures are resolved by unfolding in time to Dynamic Bayesian networks. The approach enables probabilistic inference by model averaging, i.e. it allows to predict probabilistic quantities from a set of qualitative constraints without probability assignment on the model parameters. Model averaging is performed by Monte Carlo integration techniques. The method is applied to a problem in a molecular medical context: We show how the rate of breast cancer metastasis formation can be predicted based solely on a set of qualitative biological statements about the involvement of proteins in metastatic processes.

INDEX TERMS

Probability and Statistics, Probabilistic algorithms, Uncertainty, "fuzzy", and probabilistic reasoning, Monte Carlo, Applications and Expert Knowledge-Intensive Systems, Knowledge modeling, Knowledge engineering methodologies, Biology and genetics

CITATION

Martin Stetter, Rui Chang, "Quantitative Inference by Qualitative Semantic Knowledge Mining with Bayesian Model Averaging",

*IEEE Transactions on Knowledge & Data Engineering*, vol.20, no. 12, pp. 1587-1600, December 2008, doi:10.1109/TKDE.2008.89REFERENCES

- [1] T. Bayes, “An Essay Towards Solving a Problem in the Doctrine of Chances,”
Philosophical Trans. Royal Soc. of London, 1763. - [2] P.J. Green, “Reversible Jump Markov Chain Monte Carlo Computation and Bayesian Model Determination,”
Biometrica, 1995.- [3] S.L. Lauritzen and D.J. Spiegelhalter, “Local Computations with Probabilities on Graphical Structures and Their Application to Expert Systems,”
J. Royal Statistical Soc., 1988.- [4] J. Pearl,
Probabilistic Reasoning in Intelligent Systems: Networks of Plausible Inference. Morgan Kaufmann, 1988.- [5] D. Heckerman, “A Tutorial on Learning with Bayesian Networks,” Technical Report MSR-TR-95-06, Microsoft, http://research.microsoft.com/research/pubs view.aspx?msr_ tr_id=MSR-TR-95-06, 1996.
- [6] D. Heckerman, “Learning Bayesian Networks: The Combination of Knowledge and Statistical Data,”
Proc. KDD Workshop, 1994.- [7] N. Friedman and M. Goldszmidt, “Learning Bayesian Networks with Local Structure,”
Learning in Graphical Models, 1999.- [8] E. Neufeld, “A Probabilistic Commonsense Reasoner,”
Int'l J. Intelligent Systems, 1990.- [9] M.J. Druzdzel and L.C. van der Gaag, “Elicitation of Probabilities for Belief Networks: Combining Qualitative and Quantitative Information,”
Proc. 11th Conf. Uncertainty in Artificial Intelligence (UAI), 1995.- [10] C.-L. Liu and M.P. Wellman, “Using Qualitative Relationships for Bounding Probability Distributions,”
Proc. 14th Conf. Uncertainty in Artificial Intelligence (UAI '98), pp. 346-353, 1998.- [11] S. Renooij, S. Parsons, and L.C. van der Gaag, “Context-Specific Sign-Propagation in Qualitative Probabilistic Networks,”
Proc. 17th Int'l Joint Conf. Artificial Intelligence (IJCAI '01), pp. 667-672, 2001.- [12] M.J. Druzdzel and L.C. van der Gaag, “Building Probabilistic Networks: Where Do the Numbers Come From?”
IEEE Trans. Knowledge and Data Eng., vol. 12, 2000.- [13] S. Renooij, “Qualitative Approaches to Quantifying Probabilistic Networks,” PhD dissertation, Universiteit Utrecht, 2001.
- [14] M.P. Wellman, “Fundamental Concepts of Qualitative Probabilistic Networks,”
Artificial Intelligence, 1990.- [15] S. Renooij and L.C. van der Gaag, “Decision Making in Qualitative Influence Diagrams,”
Proc. 11th Int'l FLAIRS Conf. (FLAIRS), 1998.- [16] Y. Kang, P.M. Siegel, W.P. Shu, M. Drobnjak, S.M. Kakonen, C. Cordn-Cardo, T.A. Guise, and J. Massagu, “A Multigenic Program Mediating Breast Cancer Metastasis to Bone,”
Cancer Cell, 2003.- [17] G.R. Mundy, “Metastasis to Bone: Causes, Consequences and Therapeutic Opportunities,”
Nature Rev. Cancer, 2002.- [18] A.B. Roberts and M.B. Sporn, “The Transforming Growth Factor-Betas,”
Peptide Growth Factors and Their Receptors, 1990.- [19] G.R. Grotendorst, H. Okochi, and N. Hayashi, “A Novel Transforming Growth Factor Beta Response Element Controls the Expression of the Connective Tissue Growth Factor Gene,”
Cell Growth & Differentiation, 1996.- [20] Y. Morinaga, N. Fujita, K. Ohishi, and T. Tsuruo, “Stimulation of Interleukin-11 Produced from Osteoblast-Like Cells by Transforming Growth Factor-Beta and Tumor Cell Factors,”
Int'l J. Cancer, 1997.- [21] A. Muller, B. Homey, H. Soto, N. Ge, D. Carton, M.E. Buchanan, T. McClanahan, E. Murphy, W. Yuan, and S.N. Wagner, “Involvement of Chemokine Receptors in Breast Cancer Metastasis,”
Nature, 2001.- [22] R.S. Taichman, C. Cooper, E.T. Keller, K.J. Pienta, N.S. Taichman, and L.K. McCauley, “Use of the Stromal Cell-Derived Factor-1/CXCR4 Pathway in Prostate Cancer Metastasis to Bone,”
Cancer Research, 2002.- [23] F.J. Giordano, P. Ping, M.D. McKiman, S. Nozaki, A.N. DeMaria, W.H. Dillmann, O. Mathieu-Costello, and H.K. Hammond, “Intracoronary Gene Transfer of Fibroblast Growth Factor-5 Increases Blood Flow and Contractile Function in an Ischemic Region of the Heart,”
Nature Medicine, 1996.- [24] T. Dean and K. Kanazawa, “A Model for Reasoning about Persistence and Causation,”
Artificial Intelligence, 1989.- [25] K. Murphy, “Dynamic Bayesian Networks: Representation, Inference and Learning,” PhD dissertation, Univ. of California, Berkeley, 2002.
- [26] M.A. Tanner,
Tools for Statistical Inference. Springer-Verlag, 1996.- [27] C.P. Robert,
Monte Carlo Statistical Methods. Springer-Verlag, 2004.- [28] S. Geman and D. Geman, “Stochastic Relaxation, Gibbs Distribution and Bayesian Restoration of Images,”
IEEE Trans. Pattern Analysis and Machine Intelligence, vol. 6, 1984.- [29] A. Gelman and D.B. Rubin, “Inferences from Iterative Simulation Using Multiple Sequences (with Discussion),”
Statistical Science, 1992.- [30] A.E. Gelfand and A.F.M. Smith, “Sampling-Based Approaches to Calculating Marginal Densities,”
J. Am. Statistical Assoc., 1990.- [31] A.F.M. Smith and G.O. Roberts, “Bayesian Computation via the Gibbs Sampler and Related Markov Chain Monte-Carlo Methods (with Discussion),”
J. Royal Statistical Soc., 1993.- [32] N. Metropolis and S. Ulam, “The Monte Carlo Method,”
J. Am. Statistical Assoc., 1949.- [33] J.A. Hoeting, “Methodology for Bayesian Model Averaging: An Update,” Dept. of Statistics, Colorado State Univ., 2002.
- [34] F.M. Liang, Y. Troung, and W.H. Wong, “Automatic Bayesian Model Averaging for Linear Regression and Applications in Bayesian Curve Fitting,”
Statistica Sinica, 2001.- [35] C. Fernández, E. Ley, and M.F. Steel, “Benchmark Priors for Bayesian Model Averaging,”
J. Econometrics, 2001.- [36] D. Dash and G.F. Cooper, “Model Averaging for Prediction with Discrete Bayesian Networks,”
J. Machine Learning Research, 2004.- [37] D.B. Dunson and B. Neelon, “Bayesian Inference on Order Constrained Parameters in Generalized Linear Models,”
Biometrics, 2003.- [38] H. Daumé III and D. Marcu, “A Bayesian Model for Supervised Clustering with the Dirichlet Process Prior,”
J. Machine Learning Research, 2006.- [39] J.A. Hoeting, D. Madigan, A.E. Raftery, and C.T. Volinsky, “Bayesian Model Averaging: A Tutorial,”
Statistical Science, 1999.- [41] J. Cerquides and R. López de Màntaras, “Knowledge Discovery with Qualitative Influences and Synergies,”
Proc. Second European Symp. Principles of Data Mining and Knowledge Discovery (PKDD), 1998.- [42] A.E. Raftery and D. Madigan, “Bayesian Model Averaging for Linear Regression Models,”
J. Am. Statistical Assoc., 1997.- [43] P. Dellaportas, J.J. Forster, and I. Ntzoufras, “On Bayesian Model and Variable Selection Using MCMC,”
Statistics and Computing, 2002.- [44] D. Madigan and J. York, “Bayesian Graphical Models for Discrete Data,”
Int'l Statistical Rev., 1995. |