据科技博客Gizmodo文章，现在，纽约市消防局利用数据挖掘法来预测哪些城市的建筑物发生火灾的风险最高。通过这个方法，我们可以获取到元数据。
纽约市消防局的官员使用的是60面算法，来确定哪个建筑物受火灾威胁的可能性最大，并快速追踪火灾风险高的建筑物情况。一般来说，旧的、闲置的，或者贫民区的房屋结构更容易发生火灾。有了数据挖掘项目的辅助，高危建筑物可以得到最高重视。除此之外，建筑物的数据挖掘对象是随机抽取的，其中学校和图书馆会受到格外重视。
波士顿等城市也采取了类似的方法进行数据统计。例如，波士顿对投诉电话、安全记录和税收情况等公共信息进行分析，从而确定警察巡逻的重点地区。
尽管纽约市消防局的数据统计项目尚未完成，还无法提供足够的数据来测量结果。负责风险管理的助理署长杰夫·罗斯在接受《华尔街日报(博客,微博)》采访时说道：“（这个项目）最终会让我们看到火灾发生次数减少了，火灾的严重程度也会降低。”
百度新闻与新华网国际频道合作稿件，转载请注明出处。
The New York City Fire Department is using data mining to predict which of the city"s buildings are at highest risk of catching fire. Now that"s metadata we can get behind.
FDNY officials are using a 60-facet algorithm to determine which department-inspected buildings pose the greatest fire threat, fast-tracking fire inspectors to the riskiest buildings. Structures that are old, vacant, or located in poor neighborhoods are generally at a higher risk, and thanks to the data-driven program, those structures will receive attention first. Prior to this, buildings were inspected essentially at random, with schools and libraries receiving extra attention.
Other cities have had success with similar data-driven strategies: in Boston, for example, public information including complaint calls, safety records, and tax collections are analyzed to determine where police patrols should focus.
While the FDNY data program hasn"t been in place long enough to measure results, Assistant Commissioner for Management Initiatives Jeff Roth tells the Wall Street Journal, "ultimately, we should see the number of fires go down [. . . ] and fires should become less severe."
