Generalization ability of majority vote point classifiers for motor fault diagnosis

Recently a class of classifiers named Majority Vote Point (MVP) classifiers on account of lower VC dimension were shown to give better generalization performance than linear classifiers. As the work is recent, this paper first presents a brief overview on the MVP classifier, and discusses the unique properties, with pros and cons of the classifier. Further, the paper with help of a case study, presents effective MVP classifier based solutions for bearing fault diagnosis in induction motor(s). The case study is found to clearly re-validate the effectiveness of MVP classifier in machine fault diagnosis problems.