Study on Hybrid-Weight for Feature Attribute in Naïve Bayesian Classifier
Bao-En Guo, Haitao Liu, Chao Geng · 2014
In this paper, a novel naïve Bayesian classifier based on the hybrid-weight feature attributes (short of "NBCHWFA") is proposed. NBCHWFA arranges a hybrid weight for each feature attribute by merging the effectiveness of feature attribute on classification and the dependence between feature attribute and class attribute. In order to demonstrate the feasibility and effectiveness of proposed NBCHWFA, we experimentally compare our method with standard naïve Bayesian classifier (NBC), NBC with gain ratio weight (NBCGR), and NBC with correlation coefficient weight (NBCCC) on 10 UCI datasets. And, a statistical analysis is also given. The final results show that NBCHWFA can obtain the statistically best classification accuracy.