STABLE AND FALL RECOGNITION IN ELDERLY USING PIGEON HOLE DATA REDUCTION AND OPTIMIZATION TECHNIQUE AND MODIFIED K-NN CLASSIFIER
Remote health care monitoring is a technology which enables monitoring of person outside of conventional clinical settings i.e. in the home, which may increase access to care and decrease healthcare delivery costs. UNICEF says about 50% Fall accident in the elderly person is a risk and it is increasing. Statistics indicates that one out of three people over the age of 65 will fall. Various fall detection solutions have been proposed to detect fall. In this paper, it is proposed that the data from different subjects are classified into safe and danger using a best computational classifier which can be given as an input to an automated device. The classifier is to classify the simulated data sets of elderly fall. The dataset of simulated for all the types of subjects of elderly fall was provided by a wearable dry electrode attire which is tied on the abdomen for elderly. Using a machine learning algorithm, the algorithm classifies the all the data sets into Stable and Fall for elderly . Based on the classification, it can notify the care taker or remote assistance about the them whether they are in Safe condition or not.