West Java is one of the regions in Indonesia that is prone to various natural disasters such as earthquakes, floods, and landslides. These disasters are frequent and difficult to predict, such as the tornado that hit Rancaekek, Bandung on February 2, 2024, which caused significant damage. According to data from the West Java Regional Disaster Management Agency (BPBD), this disaster resulted in many damaged buildings and injuries. An early warning system is essential to reduce the impact of disasters. This study aims to place early warning facilities based on village data in West Java using the Naive Bayes method. The method used in this study is a data mining approach to extract patterns and valuable information from data that will be used in strategic decision-making related to the placement of early warning facilities. The data used was obtained from the West Java government's open data site, which includes attributes such as codes and names of provinces, districts, sub-districts, villages/sub-districts, as well as the availability status of disaster mitigation facilities. The results of the study show that many areas in West Java still do not have adequate early warning facilities. The use of Naive Bayes' algorithm aids in data classification and provides insights into the placement of more effective early warning facilities. The implication of this study is the need for more serious and coordinated efforts from the government, non-governmental organizations, and the community to increase the availability of disaster mitigation facilities in West Java.