Intelligent System for Internet of Things-Based Building Fire Safety with Naive Bayes Algorithm

Abstract

Population growth is increasing every year. Population growth causes an increase in population density in a country. The largest population density is in urban areas. Fires in a city with a high population density will potentially cause greater damage. Material and non-material losses due to fire can be caused by not functioning maximally early warning systems, especially fire detection. In addition, other factors, such as system errors in detecting fires, can potentially cause fires. This research aims to build an intelligent system that can minimize building fire detection errors to reduce user material losses. The intelligent system can classify fire potential into four classifications, namely ”very dangerous,” ”dangerous,” ”alert,” and ”safe.” The method used in this research is Research and Development (R&D) with artificial intelligence using the Na¨ıve Bayes method, which has been integrated with the Internet of Things (IoT). This research shows that the Na¨ıve Bayes algorithm can be used to classify fire potential, proven by the overall system testing accuracy of 93.33% with an error of 6.77%.