Kangaroo Mob Heuristic for Optimizing Features Selection in Learning the Daily Living Activities of People with Alzheimer's

Dementia is a disease that affects a large proportion of elders and the number of elders that suffer from this condition is expected to increase dramatically in the next decades. Alzheimer's accounts for approximately two thirds of the people that suffer from dementia and it is a term that describes generically the memory loss. The inability to remember has major consequences upon the daily living activities patterns of the elders that suffer from this condition and consequently the study of the daily living activities performed by the people with Alzheimer's can reduce the health care costs accordingly. The main contributions of this article are: (1) the development of a novel bio-inspired algorithm called Kangaroo Mob Optimization, (2) the development of a machine learning methodology for predicting the daily living activities based on Binary Kangaroo Mob Optimization Features Selection and Random Forest and (3) the application of the proposed methodology on four datasets from literature and on one synthetic dataset in order to demonstrate the improvement of the classification performance.