×

An improved Bayesian network structure learning algorithm based on the conditional independence test. (Chinese. English summary) Zbl 1265.68178

Summary: To improve the Bayesian network structure learning algorithm based on the conditional independence test, the paper proposes an algorithm that uses mutual information between node \(x\) and \(y\). The algorithm can reduce triangulated cliques and the probability of a cyclic route. The network structure produced by the algorithm is closer to a solution. Experimental results show that the algorithm is effective and feasible.

MSC:

68T05 Learning and adaptive systems in artificial intelligence
PDFBibTeX XMLCite
Full Text: DOI