Machine Learning Methodologies for predicting Neurological disease using Behavioral Activity Mining in Health Care

Abstract

The nervous system in the human body consists of the brain, spinal cord, and nerves. Together they control all the working of the body. If any part in the nervous system in the body goes wrong, that will affect entire life and activities. We cannot do regular activities the nervous system is affected. The clinical procedure plays a crucial role as this takes more time to identify and cure the disease. This paper presents noninvasive treatment activity pattern mining and investigates the feasibility of detecting neurological disease. Our body sensor networks use some frequency range to capture data for the movement. The heel-knee-movement test gives the important activity pattern movements for the patients. The body senses the captured frequency and produces massive information about human health in terms of nerve systems. So Activity Pattern Mining (APM) differs for each patient based on the amplitude information through movements. This research shows the test results whether the test taken is positive or negative with the help of extracted amplitude. We applied different algorithms to show the various classifications. The decision tree algorithm provides better results among the applied machine learning algorithms.