Voting Principle Based on Nearest kernel classifier and Naive Bayesian classifier

This paper presented a voting principle based on multiple classifiers. This voting principle was based on the naïve Bayesian classification algorithm and a new method based on nearest to class kernel classifier that was proposed. The recognition ability of each classifier to each sample is not the same. A model of each classifier was obtained by the training on the train data, which acts as basis of the voting principle. After that, They were collected to make a decision according to the majority voting. The experiment shows that the presented voting principle achieves good performance for high recognition.