Communications of the ACM
Researchers se Twitter to Predict Crime
Researchers say they can use tweets to predict as many as 25 different types of crimes.
Credit: Agence France-Presse
Tweets can be useful for predicting 19 to 25 kinds of crimes, especially for offenses such as stalking, thefts, and certain kinds of assault, if the correct analysis is applied, according to researchers at the University of Virginia.
Matthew Gerber from the Predictive Technology Lab and colleagues analyzed tweets from Chicago tagged to certain neighborhoods and the city's crime database. Then they looked forward and were able to make useful predictions about areas where certain crimes were likely to occur. Gerber says the team's algorithm learns the pattern and produces a prediction.
"This approach allows the analyst to rapidly visualize and identify areas with historically high crime concentrations," according to the research paper on the method. "Future crimes often occur in the vicinity of past crimes, making hot-spot maps a valuable crime prediction tool."
Gerber says the results are surprising, considering people rarely tweet about crimes directly, and he notes even tweets with no direct link to crimes may contain helpful information.
From Agence France-Presse
View Full Article
Abstracts Copyright © 2014 Information Inc., Bethesda, Maryland, USA
No entries found