Abstract:Bayesian network (BN), a simple graphical notation for conditional independence assertions, is promised to represent the probabilistic relationships between diseases and symptoms. Learning the structure of a Bayesian network classifier (BNC) encodes conditional independence assumption between attributes, which may deteriorate the classification performance. One major approach to mitigate the BNC's primary weakness (the attributes independence assumption) is the locally weighted approach. And this type of approach has been proved to achieve good performance for naive Bayes, a BNC with simple structure. However, we do not know whether or how effective it works for improving the performance of the complex BNC. In this paper, we first do a survey on the complex structure models for BNCs and their improvements, then carry out a systematically experimental analysis to investigate the effectiveness of locally weighted method for complex BNCs, e.g., tree-augmented naive Bayes (TAN), averaged one-dependence estimators AODE and hidden naive Bayes (HNB), measured by classification accuracy (ACC) and the area under the ROC curve ranking (AUC). Experiments and comparisons on 36 benchmark data sets collected from University of California, Irvine (UCI) in Weka system demonstrate that locally weighting technologies just slightly outperforms unweighted complex BNCs on ACC and AUC. In other words, although locally weighting could significantly improve the performance of NB (a BNC with simple structure), it could not work well on BNCs with complex structures. This is because the performance improvements of BNCs are attributed to their structures not the locally weighting.

Not so naive Bayesian classification

Not So Naive Bayes: Aggregating One-Dependence Estimators

Improving the Naïve Bayes Classifier

Locally Weighted Learning: How and when Does It Work in Bayesian Networks?

A Technique For Improving The Performance Of Naive Bayes Text Classification

Averaged One-Dependence Estimators: Preliminary Results.

A Smart Design for the TAN Classifier

Improving Tree Augmented Naive Bayes for Class Probability Estimation

Selective Augmented Bayesian Network Classifiers Based on Rough Set Theory

Adjusting Dependence Relations for Semi-Lazy TAN Classifiers

Survey of Improving Naive Bayes for Classification

A Double Layer Bayesian Classifier

Naive Bayes Text Classifiers: a Locally Weighted Learning Approach

Attribute Weighted Naive Bayesian Classification Algorithm

ON BAYESIAN CLASSIFIER OPTIMISATION BASED ON BAT ALGORITHM

A Study of AdaBoost with Naive Bayesian Classifiers: Weakness and Improvement.

Lazy Bayesian Rules: A Lazy Semi-Naive Bayesian Learning Technique Competitive to Boosting Decision Trees

One dependence augmented naive bayes

An attributes weighted Naïve Bayesian Classifier

Lazy Learning Based Double Layer Nave Bayesian Classifier

Efficient heuristics for learning scalable Bayesian network classifier from labeled and unlabeled data