New York City Used Relevant Data Mining to Reduce Incidents of Fire The New York City Fire Department is applying data mining technique to assess fire risks in buildings. Data mining techniques are helping the New York City Fire Department to assess fire risks in buildings, says a recent report. New York City has about a million buildings and each year about 3,000 of them experience a major fire. It became necessary for the department to be able to predict which of the city's buildings are at highest risk of catching fire. The Fire Department has begun running predictive computer models to pinpoint which buildings are in danger of burning down so that preventive measures can be taken well in advance. Identifying Fire Prone Buildings Analysts at the fire department say that there are some characteristics that help to identify fire-prone buildings: •
The age of the building
•
The number and location of sprinklers
•
The Presence of Elevators
•
If, there are electrical issues
•
Building that are vacant or unguarded
The department also says that buildings located in low-income areas of the city are at greater risk. Data Mining to Build Fire Risk Score The New York Fire Department compiled a prediction model based on 60 different factors to identify risk prone buildings using SAS statistical analysis and predictive modeling tools. The algorithm assigns each one of the city’s 330,000 inspectable buildings with a risk score. Based on the score, the fire officers go on weekly
www.managedoutsource.com
18006702809