Machine Learning Classification to Detect Unattended Child in Vehicle Using Sensor Signal: A Review

A significant number of children die each year in the United States and around the world as a result of being left in hot vehicles. Numerous studies aimed at reducing the number of unattended children in vehicles have employed a variety of strategies. The majority of studies use sensors to detect unattended children, while only a few integrate machine learning with the sensors. The efficacy of a sensor's system is improved by machine learning. This paper reviews the implementation of machine learning classification in child detection systems and reviews the research conducted to detect unattended children. For the majority of the research, the machine learning algorithms SVM, KNN, and Random Forest effectively classified the occupants into a few classifications with accuracies greater than 90%.