Naïve Bayes Associative classification of mammographic data

In this paper we focus on a new model, named ANB (Associative Naïve Bayes) model. ANB model extend the modeling flexibility of well known Naïve Bayes (NB) models by introducing rules generated by associative classifier. The model consists of two layers: an input layer and an internal layer. We propose an associative classifier algorithm (AAC), relaxing the condition of independence of attributes in NB, for generating rules and learning network parameter and a simple algorithm for training ANB models in the context of classification. Experimental results show that the learned models can significantly improve classification accuracy as compared to NB.