|
Abstract
|
|
|---|---|
| Multi-dimensional Bayesian network classifiers (MBCs) are Bayesian network classifiers especially designed to solve multidimensional classification problems, where each instance in the data set has to be assigned to one or more class variables. In this paper, we introduce a new method for learning MBCs from data basically based on determining the Markov blanket around each class variable using the HITON algorithm. Our method is applied to the human immunodeficiency virus (HIV) protease inhibitor prediction problem. The experimental study showed promising results in terms of classification accuracy, and we gained insight from the learned MBC structure into the different possible interactions among protease inhibitors and resistance mutations. | |
|
International
|
Si |
|
Congress
|
13th Conference on Artificial Intelligence in Medicine (AIME?11) |
|
|
960 |
|
Place
|
Bled, Slovenia |
|
Reviewers
|
Si |
|
ISBN/ISSN
|
|
|
|
|
|
Start Date
|
02/07/2011 |
|
End Date
|
|
|
From page
|
29 |
|
To page
|
40 |
|
|
Proceedings of Probabilistic Problem Solving in BioMedicine Workshop at 13th Conference on Artificial Intelligence in Medicine (AIME?11) |