An Empirical Evaluation of Learning Models for the Classification of Fall Detection Dataset
Monika Monika, Sakshi Sinha · 2024
This chapter aims at studying different classification algorithms and finding and comparing their accuracy in identifying the fall of an elderly person, with accuracy. Falls are considered a very dangerous health problem in public, especially for elderly people. It can be life-threatening for people in fall risk groups meaning the people of elderly age. For this study, a dataset which is based on the falls of elderly people have been used. There are seven different values of the vitals given for an elderly person which are related to the functioning of heart and brain along with the activities they are performing at that time. With the help of this data the outcome of that activity is being categorized in a fall or no fall scenario. The threshold value of each parameter being used for this classification has been obtained after intensive study from various medical records and sites. The experiment studies different algorithms for finding the best one and the results show that the Random Forest Model accurately classifies with 99.1% accuracy as compared to other models.