Technology has transformed city living – you can order an Uber with a tap, pay for a coffee with an NFC touch, jump onto a wireless charging point, and track your steps, so you know how far you’ve walked and how many calories you’ve burned.
Data analytics and cloud computing power a lot of the glossy consumer apps that have slotted into our daily lives quickly and effortlessly.
But there’s more to the so-called ‘smart city’ concept than making our lives convenient. By taking in information like real-time facial recognition, CCTV video link-ups, and license plate scanning, authorities all around the world are waking up to the idea that data analytics and cloud computing can be used to tackle weightier issues like crime as well.
In Rio de Janeiro, for example, the city’s Operation Centre has 500 people working in a state of the art data sharing facility for just this purpose. In Singapore, the introduction of smart mapping technology is pioneering crime forecasts.
Over in Detroit, Project Green Light shares video and images between police departments in real time, while in Cape Town, one of the world’s most violent cities, a system called ShotSpotter maps shooting patterns for later analysis.
London is also going high-tech. A number of departments are using new tools to more accurately plot the locations of violent crimes. The data, drawn from an array of sources, is helping emergency services detect patterns and identify previously unknown crime hotspots.
Most recently, the Mayor’s Office for Policing and Crime (MOPAC) has been working with the Greater London Authority’s Strategic Crime Analysis team in a data analysis project, using a secure web portal called SafeStats.
Over a number of years, the Strategic Crime Analysis team has collated crime reports, analysing and sharing over twenty million records. These are made up of crime and safety data from the police, ambulance service, fire brigade and transport authorities, providing a significant data pool.
The new initiative is based around taking this data, and combining it with more general information given to hospital staff – for example, during the sign-in process at an Accident and Emergency unit. This is based on the belief that a large percentage of crime isn’t reported officially (the government claims only twenty-three percent of all emergency department attendees who receive medical treatment for injuries sustained from violent assaults report it to the police).
Combining the detailed information that is already entered into the SafeStats system from reported crime, with a wider but less detailed data set based on hospital reports, law enforcement agencies can conduct comparative analysis. They claim this has the potential to identify previously unknown hotspots of violence – with a particular relevance to youth, domestic and gang violence.
Data from 25 hospitals can now be cross-referenced with 15 million SafeStats records.
One challenge the project faces is producing accurate location data based on ‘freetext’ data entries – i.e., due to the non-specific nature of the information given to hospital staff by those suffering injury, they have to work with what they are given. Rather than say, selecting a specific geo-tagged location from a list of predetermined options, reports are likely to be entered as something more vague, such as ‘McDonald’s on King’s Road’
To solve this data capture problem, software has been developed that recognises the address elements present in any freetext string (such as a street, town, shop, bar or school). Then, based on the combination of the elements present, it prioritises the likelihood of an accurate location match. In-house lookups, online geocoders, and spatial lookups are then utilised to assign locations at varying geographic levels.
Through this process, the data received at hospitals is collated, standardised, address-matched and published in minutes. Coupled with the records from other agencies, it can provide an up-to-date picture of incidents in the city.
Data from 25 hospitals can now be cross referenced with 15 million SafeStats records from the Metropolitan Police, London Ambulance Service, Transport for London, British Transport Police, and the London Fire Brigade.
From this, a pan-London picture can be created, which allows law enforcement authorities to access non-police data to identify potential crimes, incidents and hotspots. Based on that, local policing tactics are updated accordingly.
“Keeping Londoners safe is the Mayor’s top priority, and it’s vital that the emergency services, local authorities, transport bodies and hospitals work together to do everything possible to drive down crime in the capital,” says Sophie Linden, The Deputy Mayor for Policing And Crime.
“By sharing data and intelligence through the SafeStats portal, police can target their resources depending on local need. In addition, as we continue to expand our work with emergency departments, we can develop our understanding of crime, improve the police response and help keep communities safe.”
MOPAC claims that this is the first time anything like this has been successfully implemented on a city-wide scale, and the word is spreading. The findings are being presented to authorities in New Zealand, Australia and the United States, who are looking to replicate the success London has had with bringing data analytics into the fight against crime.