Novel RVM Approach to Structuring and Classifying Epidemic Outbreak Data

Abstract

Classifying this indefinite big data, is computationally intensive as a large amount of data is related with an existential probability of undefined or undetermined values of raw data. Classifying is hindered by a large amount of data from various sources. RVM, a Bayesian formulation of the linear model both for classification and regression, has lately involved a lot of interest in the research community. The paper aims at learning kernelized RVM classifier to evaluate Ebola virus outbreak, using generalization error, intra class separability, missing probability Pi is compared to SVM.RVM relevance impact with other epidemic diseases of Ebola Virus is also compared.