A Minority Report Future?

Have you watched Minority Report? You know, an utopian society where crime is more or less non-existent – simply because “criminals” are caught before a crime is even committed, based on a system that claims to be able to figure out the thoughts and intent of a particular person. Hence, making the maxim “prevention is better than cure” lived out to the fullest. A team at the University of Trento, Italy, claimed that they might be on to something similar, being able to work out whether an area will feature a low or high crime rate – although the accuracy level remains at 70%.

The researchers relied on information that has been collected by Telefonica Digital’s Smartsteps database, where it will supposedly collect anonymized and aggregated mobile network data, letting researchers spot footfall trends by time, gender, and age. The dataset will then break down the capital into 124,119 cells, enabling analysts to check out the footfall in a particular region throughout the day. Apart from that, data from the London Borough Profiles Dataset will be used, and these include vital statistics on the housing market, political affiliation, transportation, homelessness and life expectancy, alongside information concerning individual crimes from the London Criminal Cases database.

All of those are a potent mix, and across a long period of time, it might just make the prediction all the more accurate. The researchers mentioned, “Usage of human behavioral data (at a daily and monthly scale) significantly improves prediction accuracy when compared to using rich statistical data about a borough’s population (households census, demographics, migrant population, ethnicity, language, employment,etc…). The borough profiles data provides a fairly detailed view of the nature and living conditions of a particular area in a city, yet it is expensive and effort-consuming to collect. Hence, this type of data is typically updated with low frequency (e.g. every few years). Human behavioral data derived from mobile network activity and demographics, though less comprehensive than borough profiles, provides signicantly finer temporal and spatial resolution.”

You can read more on their research here.

Edwin Kee
Categories: Computers
Tags: Science