Twitter has proven itself to be an important tool for communication when sudden calamities strike. During the 2011 London riots, for instance, information on broken windows and burning cars was able to make it to the public quicker than people were able to report it to police by as much as an hour. Using a series of algorithms, researchers were able to determine where people were gathering and confirm rumors of where riots were likely to take place a new study has shown.
Researchers from Cardiff University were able to demonstrate how their algorithms could scan through posts and hashtags and reliably predict possibly dangerous occurrences. While London police and other police districts here in the US have been discussing budget cuts within the near future, some are now saying that Twitter could possibly prove to be the most valuable asset in preventing crimes from happening.
Dr. Pete Burnap, Co-author of the study says,
“In this research we show that online social media are becoming the go-to place to report observations of everyday occurrences — including social disorder and terrestrial criminal activity. We will never replace traditional policing resource on the ground but we have demonstrated that this research could augment existing intelligence gathering and draw on new technologies to support more established policing methods.”
One of the first tests of these algorithms occurred in 2015, during a time of civil unrest in the Middle East. The data collected included 40 million tweets from 18 million individual user accounts and generated 425,000 unique hashtags. These came together to give researchers an accurate picture of not only isolated crimes but also updates on transportation and the weather.
Of course, researchers also say that their system is not perfect. Character limitations, misspellings, abbreviations and slang words often cause their algorithms to miss certain tweets which may offer them a broader picture of what is occurring within a specific region.
Some police have already adopted software in order to track social media. Software packages such as MediaSonar, Geofeedia, and X1 Social Discovery have been used to track events in certain geographic regions and have already been utilized by police departments to track the movements of certain groups such as Black Lives Matter, reports Forensic Magazine. According to Geofeedia, more than 500 police agencies have already adopted their technology alone.