Applying Machine Learning Algorithm in Fall Detection Monitoring System

Fall is a major health hazard for the elders when they live independently. Approximately a third of those aged 65 years and over fall each year. An automatic fall detector ensures the best possible chance of a full recovery following a fall. This paper presents new algorithm able to learn, classify and identify falls from data obtained by a multi-sensor monitoring system. The system, that uses a web cam and a heart rate sensor, is based on machine learning and data classification using decision trees. Our solution shows a satisfactory performance and gives interesting results.