Combining one class classification models for avian influenza outbreaks

The prediction of avian influenza outbreak animal cases is a genuine one class classification issue because the real world outliers are impractical to obtain. In this paper, a new combining one class classification method has been presented and illustrated on the avian influenza outbreak dataset. The presented combining methods outperform the previous combining methods both on the original avian influenza outbreak dataset and dimension reduction one. The new one classification combining model can be adapted to the warning surveillance purpose and proved to be practical on the avian influenza outbreak prediction tasks.