Graph-Based Analysis of Nasopharyngeal Carcinoma with Bayesian Network Learning Methods

Authors:
Alex Aussem;Sergio Rodrigues De Morais;Marilys Corbex;Joël Favrel
Affiliations:
University of Lyon, LIESP, Villeurbanne, France F-69622;University of Lyon, LIESP, Villeurbanne, France F-69622;International Agency for Research on Cancer (IARC), Lyon Cedex 08, France F-69280;University of Lyon, LIESP, Villeurbanne, France F-69622
Venue:
GbRPR '09 Proceedings of the 7th IAPR-TC-15 International Workshop on Graph-Based Representations in Pattern Recognition
Year:
2009

Citing 9
Cited 0

Probabilistic reasoning in intelligent systems: networks of plausible inference

Probabilistic reasoning in intelligent systems: networks of plausible inference
Large-Sample Learning of Bayesian Networks is NP-Hard

The Journal of Machine Learning Research
The max-min hill-climbing Bayesian network structure learning algorithm

Machine Learning
Growing Bayesian network models of gene networks from seed genes

Bioinformatics
Learning Bayesian Networks

Learning Bayesian Networks
Consistent Feature Selection for Pattern Recognition in Polynomial Time

The Journal of Machine Learning Research
Towards scalable and data efficient learning of Markov boundaries

International Journal of Approximate Reasoning
A Novel Scalable and Data Efficient Feature Subset Selection Algorithm

ECML PKDD '08 Proceedings of the European conference on Machine Learning and Knowledge Discovery in Databases - Part II
Robust Gene Selection from Microarray Data with a Novel Markov Boundary Learning Method: Application to Diabetes Analysis

ECSQARU '09 Proceedings of the 10th European Conference on Symbolic and Quantitative Approaches to Reasoning with Uncertainty

Quantified Score

Hi-index	0.00

Visualization

Abstract

In this paper, we propose a new graphical framework for extracting the relevant dietary, social and environmental risk factors that are associated with an increased risk of Nasopharyngeal Carcinoma (NPC) based on a case-control epidemiologic study. This framework builds on the use of Bayesian network for representing statistical dependencies between the random variables. BN is directed acyclic graphs that models the joint multivariate probability distribution underlying the data. These graphical models are highly attractive for their ability to describe complex probabilistic interactions between variables. The graph provides a statistical profile of the recruited population and meanwhile help identify a subset of features that are most relevant for probabilistic classification of the NPC. We report experiment results with the NPC case-study data using a novel constraint-based BN structure learning algorithm. We show how the DAG provides an improved comprehension of NPC etiology. Our findings are compared with the risk factors that were suggested in the recent literature in cancerology.