TAN Classifiers Based on Decomposable Distributions

Abstract

Abstract. In this paper we present several Bayesian algorithms for learning Tree Augmented Naive Bayes (TAN) models. We extend the results in Meila & Jaakkola (2000a) to TANs by proving that accepting a prior decomposable distribution over TAN's, we can compute the exact Bayesian model averaging over TAN structures and parameters in polynomial time. Furthermore, we prove that the k-maximum a posteriori (MAP) TAN structures can also be computed in polynomial time. We use these results to correct minor errors in Meila & Jaakkola (2000a) and to construct several TAN based classifiers. We show that these classifiers provide consistently better predictions over Irvine datasets and artificially generated data than TAN based classifiers proposed in the literature.

A decision support methodology to enhance the competitiveness of the Turkish automotive industry

“…Several search algorithms such as simulated annealing algorithms, genetic algorithms, and Three Augmented Naïve Bayes (TAN) algorithms (Cerquides and Mantaras, 2005) have been developed for this purpose (Wu, 2010;Hruschka and Ebecken, 2007;Baesens et al, 2004;). The knowledge-based approach, on the other hand, uses the causal knowledge of domain experts to construct networks.…”

An Improved De-noising Algorithm for Bayesian Network Classifiers Parameter Learning

“…Step 5: compute the weight of sample ‫ݏ݊ܫ‬ ∈ ‫ܦ‬ based on formula (4) Step 6: compute the weight of sample ‫ݏ݊ܫ‬ ∈ ‫ܦ‬ based on formula (5) Step 7: make ‫,݅‪ሺ‬ݐܦ‬ ݆ሻ = ‫ݐܦ‬ሺ݅ሻ…”

“…However, it is not easy to generate the optimal Bayesian network in practice [2][3]. Therefore, the training of restricted Bayesian network classifiers has become a very active research field in recent years [4][5][6][7].…”

An Ensemble of Fine-Tuned Heterogeneous Bayesian Classifiers

“…The tables show that the average classification accuracy of EBN-20 is 71.56% which is better than the average accuracy of any of its 20 constituent classifiers, except for FTTAN-TAN and FTTAN-K2. The result not a surprising because TAN search and K2 search algorithms have exhibited excellent performance in data mining [25] [26] and the fine tuning process makes them even better. The degradation of EBN-20 average accuracy is probably because EBN-20 combines finetuned and non-fine-tuned classifiers, which reduces diversity, as the constituent classifiers are not very different classifiers.…”