Abstract
More and more applications have come with large data sets in the past decade. However, existing algorithms cannot guarantee to scale well on large data. Averaged n-Dependence Estimators (AnDE) allows for flexible learning from out-of-core data, by varying the value of n (number of super parents). Hence, AnDE is especially appropriate for large data learning. In this paper, we propose a sample-based attribute selection technique for AnDE. It needs one more pass through the training data, in which a multitude of approximate AnDE models are built and efficiently assessed by leave-one-out cross validation. The use of a sample reduces the training time. Experiments on 15 large data sets demonstrate that the proposed technique significantly reduces AnDE's error at the cost of a modest increase in training time. This efficient and scalable out-of-core approach delivers superior or comparable performance to typical in-core Bayesian network classifiers.
Funding Information
  • Australian Research Council (DP140100087)
  • Asian Office of Aerospace Research and Development
  • Air Force Office of Scientific Research (FA2386-15-1-4007)
  • National Natural Science Foundation of China (61202135)
  • Natural Science Foundation of Jiangsu (BK20130735)
  • Natural Science Foundation of Jiangsu Higher Education Institutions of China (14KJB520019, 13KJB520011, 13KJB520013)
  • Key Laboratory of Symbolic Computation and Knowledge Engineering of Ministry of Education
  • Development of Jiangsu Higher Education Institutions
  • Monash e-Research Center and eSolutions-Research
  • HPC
  • LIEF
  • NCI National Facility in Canberra
  • Australian Commonwealth Government