Big Data meets Minority Report: Predicting crimes in real time
By Nathan McGregor, VP Australia and NZ, Hitachi Data Systems
The way we connect with the world is rapidly evolving. Businesses, individuals and governments must constantly innovate to keep pace with global and local trends, to manage disruptions and capture market opportunities.
The biggest opportunity lies with the Internet of Things (IoT), in what many are calling “the next Industrial Revolution”. In this highly pressurised environment, organisations not only need to innovate faster, make smarter decisions and build more successful businesses, but also assist the world around us.
The answer lies with data. From smartphones to wind turbines, and trains to farms, data is everywhere thanks to IoT and connected devices. Although IDC estimated the big data market in Asia Pacific to be worth US$1.2 billion in 2014, it’s growing quickly with a CAGR of 32 per cent through 2018 when the market will reach US$3.6 billion.
The real value of data comes from identifying insights through analytics. Although analytics’ solutions are not new, with big data we now have a much wider array of data sources that can drive valuable insights.
New data, new technology challenges and new opportunities – this is the world of big data.
Changing how we police
The world’s population has doubled over the last 50 years. We will reach around 8.8 billion people by 2030, with a majority of those living in urban areas by 2050.
According to a report by Navigant Research, investment in smart city technologies in Asia Pacific will amount to US$63.4 billion by 2023, with more than 100 smart cities planned in India alone.
These and other significant shifts in demographics will inevitably create challenges in sectors such as healthcare, innovations for smart city planning, sustainable energy, improved transportation and more effective infrastructure solutions.
New technologies for example, are delivering unprecedented volumes of data to public safety officials. This information originates from a variety of sources, such as video surveillance cameras, that can aid crime investigations. Similarly, sensors on traffic systems, public transportation and other critical infrastructure have the potential to uncover equipment malfunctions, unsafe conditions and gridlock.
However, this flood of information is more than most public safety organisations can effectively assess with human resources alone. To act swiftly in the face of an emergency, public safety officials require more than fast and easy access to data from their surveillance systems. They need data compiled, analysed and delivered as actionable information in real-time – right at their fingertips.
Efficiently viewing and analysing data from numerous sources requires a new generation of tools. These tools run on top of existing systems and have the power to extend their reach and usefulness while providing deep insights. They simplify the user experience by placing data from disparate systems on a common platform and associating metadata with millions of evidence clips that can be retrieved for further analysis.
Hitachi Data Systems recently announced the newly enhanced Hitachi Visualization Suite 4.5 (HVS) with predictive crime analytics, called Hitachi Visualization Predictive Crime Analytics (PCA).
HVS is a hybrid cloud-based platform that integrates disparate data and video assets from public safety systems – such as license plate readers or gunshot sensors – in real time, while presenting this information geospatially, in easily digestible forms such as heat maps. HVS provides law enforcement teams with critical information to improve intelligence, enhance investigative capabilities and increase operational efficiencies.
Blending real-time event data with historical and contextual crime data from record management systems and other sources, PCA’s powerful spatial and temporal prediction algorithms aid law enforcement agencies and first responder teams assign threat levels for every part of the city. These algorithms can also be used to create threat level predictions to accurately forecast where crimes are likely to occur or additional resources required.
PCA is the first predictive policing tool that uses natural language processing for topic intensity modeling. It utilises social media networks with other public and private data feeds in real time to deliver very accurate crime predictions.
In the US, the Washington, D.C. Metropolitan Police Department uses similar solutions that enable hundreds of personnel to view integrated video feeds, sensor data and information from third-party systems. All information is overlaid onto a map of relevant areas and analytics software can tap into information from public sources, such as social media and online news, to search for correlations and anomalies.
Changing how we think
For big data to truly have an impact on society, we need to understand all the areas of our lives that it touches, and shift our thinking to a big data mind-set.
The challenge will continue to revolve around ways to identify and analyse the data that matters, uncover patterns, and use the findings to make more informed and efficient decisions.
Through innovative technologies and total solutions for sustainable urban development, big data can help us address critical global issues such as public safety, improve businesses and cities. It could even save lives.