The New York City Fire Department uses data mining to predict which of the city's 1 million buildings are most at risk of a major fire. By analyzing 60 factors such as a building's age, electrical issues, sprinkler system, and vacancy status, the department calculates a risk score for each of the city's 330,000 inspectable buildings. Fire officers are directed to inspect higher-risk buildings first based on these scores. The data-driven approach aims to reduce the number and severity of fires compared to the previous random inspection method. Other cities like Boston are also using big data to identify problem properties for targeted police visits.