Development of Integrated System of Divided Prediction Results for Disease Classification

摘要

The Integrated System of Divided Prediction Results is a system that combines the results using a multiple classifier. Divide the class that classifies per classifier into subsets to improve the accuracy of classification items and combine results of the classifiers. This provides an environment in which classification items can be easily expanded while securing higher reliability like an ensemble classifier [1]. Classification items are classified into duplicate items and unique items to obtain weights. In this process, we propose a weights combination method.