Ensemble Extreme Learning Algorithms for Alzheimer’s Disease Detection

Abstract

Alzheimer's disease has proven to be the major cause of dementia in adults, making its early detection an important research goal. We have used Ensemble ELMs (Extreme Learning Models) on the OASIS (Open Access Series of Imaging Studies) data set for Alzheimer's detection. We have explored various single layered light-weight ELM networks. This is an extension of the conference paper submitted on implementation of various ELMs to study the difference in the timing of execution for classification of Alzheimer's Disease (AD) Data. We have implemented various ensemble ELMs like Ridge, Bagging, Boosting and Negative correlation ELMs and a comparison of their performance on the same data set is provided.