Scaling-Up Bayesian Network Learning to Thousands of Variables Using Local Learning Techniques

State-of-the-art Bayesian Network learning algorithms do not scale to more than a few hundred variables; thus, they fall far short from addressing the challenges posed by the large datasets in biomedical informatics (e.g., gene expression, proteomics, or text-categorization data). In this paper, we present a BN learning algorithm, called the Max-Min Bayesian Network learning (MMBN) algorithm that can induce networks with tens of thousands of variables, or alternatively, can selectively reconstruct regions of interest if time does not permit full reconstruction. MMBN is based on a local algorithm that returns targeted areas of the network and on putting these pieces together. On a small dataset MMBN outperforms other state-of-the-art methods. Subsequently, its scalability is demonstrated by fully reconstructing from data a Bayesian Network with 10,000 variables using ordinary PC hardware. The novel algorithm pushes the envelope of Bayesian Network learning (an NP-complete problem) by about two orders of magnitude.