Probabilistic reasoning in intelligent systems: networks of plausible inference
Probabilistic reasoning in intelligent systems: networks of plausible inference
Data mining: practical machine learning tools and techniques with Java implementations
Data mining: practical machine learning tools and techniques with Java implementations
Machine Learning
Machine Learning
Time and sample efficient discovery of Markov blankets and direct causal relations
Proceedings of the ninth ACM SIGKDD international conference on Knowledge discovery and data mining
Not So Naive Bayes: Aggregating One-Dependence Estimators
Machine Learning
Statistical Comparisons of Classifiers over Multiple Data Sets
The Journal of Machine Learning Research
Towards scalable and data efficient learning of Markov boundaries
International Journal of Approximate Reasoning
The Journal of Machine Learning Research
The Journal of Machine Learning Research
Probabilistic Graphical Models: Principles and Techniques - Adaptive Computation and Machine Learning
A fuzzy intelligent approach to the classification problem in gene expression data analysis
Knowledge-Based Systems
A Method for Integrating Expert Knowledge When Learning Bayesian Networks From Data
IEEE Transactions on Systems, Man, and Cybernetics, Part B: Cybernetics
New skeleton-based approaches for Bayesian structure learning of Bayesian networks
Applied Soft Computing
Hi-index | 0.00 |
The discovery of the Markov Boundary (MB) of a target variable using observational data plays a central role in feature selection and local causal structure inference. Most existing methods previously employed for this task rely on statistical independence tests and, in consequence, do not take into account the partial evidence that a finite data set gives about the existence of this kind of probabilistic relationships among random variables. In this work, we employ a novel stochastic search method which explicitly deals with this problem by eliciting multiple alternative Markov boundaries. This technique is based on a Bayesian approach for statistical tests and on a method to score the different alternative solutions. We have also evaluated an interactive procedure for integrating domain or expert knowledge a posteriori (after the learning process), in order to simplify and enrich the set of alternative inferred MBs. In an extensive experimental evaluation we show that this method is able to find a rich and accurate set of alternative MBs which, if properly combined, provide better inferences than other state-of-the-art approaches for this task. Moreover, we think that this new kind of methods, capable of capturing the inherent uncertainty of any real data set and which allows human interventions, can make practitioners feel more confident about the extracted knowledge than fully automatic approaches.