Research Article
A Decomposition Algorithm for Learning Bayesian Networks Based on Scoring Function
Algorithm 2
Algorithm for discovering Markov boundary.
(1) Input: Data set ; Variable set ; Target variable . | (2) Initialization: . | (3) Order-0 CI test: for each variable , if is hold, then . | (4) Order-1 CI test: for each variable , if there is a variable such that | , then . | (5) Find superset of spouses: for each variable , if there is a variable | , such that , then . | (6) Find parents and children of : call the MMPC algorithm to get | . For each , if , | then . | (7) Find spouses of : for each variable , if there is a variable | and a subset , such that | and , then . | (8) Return . | (9) Output: A Markov boundary of . |
|