A robust technique for prediction of dementia using ensemble voting classifier

Abstract

Dementia is an ordinary term used to describe the decline in memory, language, problem-solving, and other cognitive abilities that significantly disrupt activities of daily living. Dementia can be caused by an assortment of diseases. In a global society where the elderly population continues to increase, dementia is becoming an increasing concern. Every three seconds, somebody in the world gets dementia. It will reach 7 8 million in 2030 and 135 million in 2050 [1]. Dementia costs more than US $ 1.3 trillion a year around the world, and that number is expected to reach US $2.8 trillion by 2030 [2]. Despite its high prevalence within the community, this mental health condition remains inadequately identified, reported, and even partially comprehended. As a result of the exponential growth of computational power, scholars have created machine learning (ML) methods for the detection and diagnosis of neurodegenerative diseases. This paper aims to accelerate the prediction of dementia at an early stage in an individual's development. Our experiment's findings reveal that the Voting Classifier outperforms other methods, with an impressive prediction accuracy of 99.9%.