At first, researchers had a tough time believing that the social media tool known as Twitter could be helpful as a crime-prediction tool. At first blush, the 140-character limit used by the Twitter application can lead to the use of slang and abbreviations and neologisms or the name for a newly coined term, word, or phrase.
In Chandler, Phoenix or other communities in Maricopa County, it is important that you consult Arizona lawyers experienced in DUI and related offenses like those associated with the Law Offices of Craig W. Penrod, P.C.
While criminals have been known to taunt law enforcement agencies using Twitter, these tweets rarely reveal the nature of their plans in advance of commission of the crimes. Thus, the hypothesis was there was little information of substance included in these brief communications.
Nevertheless, the University of Virginia's Predictive Technology Lab was able to discover that public Twitter data improved the predictions for 19 of 25 crimes that occurred in the metropolitan Chicago area, when compared with predictions based upon historical crime patterns alone. Predictions for stalking, criminal damage and gambling resulted in the biggest jump.
The experiment began with researchers collecting more than 1.5 million public tweets that were tagged with global positioning system (GPS) coordinates within the city limits of Chicago between January and March 2013. It should be noted that Twitter users have an option to enable GPS tagging for their tweets. At the same time, researchers gathered information regarding all documented crimes that occurred over that same period. The same study could have produced similar results in the Phoenix area.
Using a computer algorithm that separated the tweets into one kilometer by one kilometer neighborhoods, the content of the tweets in each neighborhood was analyzed to determine what people were tweeting about. The content was placed into categories or topics. As an example, tweets around Chicago O'Hare International Airport pertained to travel, with tweets including words like gate, plane, flight, as well as delayed. Similar data could be gathered for the neighborhoods surrounding Sky Harbor Airport.
In basic terms, the research model compared topics in a neighborhood with historical crime data from that same spot for a given month resulting in the formation of correlations between topics and crimes. Those correlations were used to predict crime in the same neighborhood in the near future. Crime date showed a cluster of thefts that occurred in a particular neighborhood and a review of the the Twitter content generated for that particular network showed words that are highly associated with theft allowing a prediction on the basis of that content.
For 19 crimes that occurred during these months in Chicago, the research model provided a more reliable prediction than the historical crime data alone. Researchers suggested that persons were tweeting about plans that correlated with illegal activity, as opposed to tweeting about the actual crimes. The algorithm used to review the Twitter content and historic crime data identified 700 Twitter topics related to criminal damage. One of the topics involved the words "united center blackhawks bulls" and so on. This scenario extrapolates far more than is revealed by the data, but it reveals a window into the predictive power of the algorithm.
From a logistical standpoint, the Twitter data and the modeling software used by researchers are available. The key is whether a law enforcement agency uses the same historical crime data as a baseline for comparison. If not, a new round of tests would have to be done to show that the addition of Twitter data still offered a predictive upgrade.
Crime prediction driven by analysis of data has the potential raise a number of civil rights concerns. Recent revelations regarding the collection of meta data from smart phone use by the National Security Agency (NSA) has resulted in concerns from many in this country and elsewhere. The methods used by the Chicago Police Department have been referred to as a high-tech method for racial profiling. Researchers participating in this study regarding Twitter said their algorithms do not target any individuals, but only gather data that is posted voluntarily to a public account.
It is important that your legal situation should be handled by expert professionals with the experience possessed by the attorneys at the Law Offices of Craig W. Penrod, P.C. Our office offers free consultations for all DUI and criminal matters. Our Arizona DUI attorneys and Arizona criminal lawyers can provide you with the experience and knowledge needed for qualified representation. An experienced Arizona DUI lawyer is vital in DUI cases and our DUI attorneys set that standard. If you're in need of a criminal defense, our Arizona criminal attorneys are ready to assist you.