A two-phase method for extracting explanatory arguments from Bayesian networks.
Sjoerd T. TimmerJohn-Jules Ch. MeyerHenry PrakkenSilja RenooijBart VerheijPublished in: Int. J. Approx. Reason. (2017)
Keyphrases
- bayesian networks
- detection method
- high accuracy
- objective function
- machine learning
- dynamic programming
- experimental evaluation
- em algorithm
- support vector machine svm
- computational cost
- pairwise
- classification method
- similarity measure
- missing data
- semi supervised
- optimization algorithm
- parameter learning
- mutual information
- edge detection
- denoising
- classification accuracy
- probabilistic model
- multiscale
- decision trees