“School of Biological Sciences”

Back to Papers Home
Back to Papers of School of Biological Sciences

Paper   IPM / Biological Sciences / 14427
School of Biological Sciences
  Title:   A novel node ordering method using L1-regularized Markov Blanket and modified likelihood reduction factor for the K2 algorithm
  Author(s): 
1.  Vahid Rezaei Tabar
2.  Farzad Eskandari
3.  Selva Salimi
  Status:   inProgress
  Journal: pattern recognition letter
  Year:  2016
  Supported by:  IPM
  Abstract:
One of the most effective structure-learning methods in Bayesian network is K2 algorithm. Because the performance of the K2 algorithm depends on node ordering, more effective node ordering inference methods are needed. In this paper, we introduce a novel node ordering method based on the L1-regularized Markov Blanket and Modified Likelihood Reduction Factor (LRF). For this purpose, based on the fact that the parent and child variables are identified by estimated Markov Blanket (MB), we first estimate the MB of a variable by using the L1-regularized Markov Blanket. We then determine the candidate parents of a variable by evaluating the conditional frequency associations using a modification LRF. In other words, we introduce a novel scoring which infers the better parent variable through the estimated MB. Then the candidate parents are used as input for the K2 algorithm. Experimental results for most of the datasets indicate that our proposed method significantly outperforms previous method.

Download TeX format
back to top
scroll left or right