In today’s world, vast amounts of data are being harvested every second. These large volumes of raw data need to be transformed into manageable information before they can be put to use for purposes such as supporting public safety. For this data to become useful, it needs to be processed, which entails storing, cleaning, and organizing the data.
Data can be structured, such as databases of names, addresses, or drivers’ license numbers, or it can be unstructured, such as social media posts, satellite imagery and video files. Unstructured data presents more challenges because it cannot be fed into relational databases that are ready for analysis. Once the data clean-up and organization are done, the data can be analyzed to extract relevant information and identify patterns and trends for a wide range of sectors, including public safety
Studying for a career in public safety
When you join the Public Safety Laurier University program online, you will gain a postgraduate qualification designed to bolster your career in the public safety sector. The Master’s of Public Safety Degree is the first of its kind and has been developed to incorporate the four pillars of Public Safety Canada.
The Master’s program is 100% online and fully flexible, so you can continue working through the 30-month part-time course.
Using data to enhance public safety
Here is a look at some of the ways data can be used to enhance public safety.
Policing and justice
Policing has traditionally been a reactive service. Officers are sent to respond to incidents that have already occurred. However, with the use of data science, that is now changing. No longer confined to the world of movies, crime prediction is now a real-world tool. Using machine learning, data analysts can drill down into large datasets to develop criminal profiling.
Predictive analytics involve reviewing historic data to identify areas with the highest rates of crime. This information can be used to forecast likely geographical hotspots for future crime. Law enforcement can then optimize the allocation of security resources, thereby reducing the likelihood of some criminal activities taking place.
A joint study by Telefonica, Columbia University and the City of New York in 2018 used data analysis to predict potential crimes in specified areas of the city. The study found that big data could be useful in predicting crime, and predictive policing has been employed in major cities such as Chicago and London as well as New York.
Big data can be used to inform decisions on sentencing for criminals. In some places, data is used to support bail and parole decisions. Analytical tools can be used to assess the likelihood of a person skipping bail or reoffending and assign a risk score to offenders. In the US, the Correctional Offender Management Profiling for Alternative Sanctions (COMPAS) system is a decision-support software package that has been used to assess defendants and offenders in terms of:
- Pre-trial release risk
- General recidivism risk
- Violent recidivism risk
It should be noted that the COMPAS system has been criticized by some groups, partly because the algorithms are not in the public domain and therefore cannot be examined and subjected to due process.
Algorithms can also be used to evaluate the needs of offenders who are due to be released from prison and recommend the most suitable type of intervention plan. For example, some offenders might need education-focused interventions in order to build a new life for themselves and find employment once they are back out in the community.
Criminal profiling involves building a picture of a potential criminal perpetrator using behavioral connections and clues. It is a time-consuming process when carried out by teams of investigators, but machine learning techniques can be used to make it far more efficient to spot patterns and generate a criminal profile.
Specialist software such as speech analysis programs can be used to identify connections between criminals or suspicious individuals, while other tools can trawl through Facebook posts and emails to highlight potentially suspicious activity and contacts.
Data mining and risk tracking for intervention
In many cities across Canada, police agencies make use of a risk-driven tracking database that collates information from a variety of sources, including schools, social work agencies and police records, to track negative behavior. The police can then make proactive interventions that can deter potential criminal activity.
It is important for police forces to have access to the software they need to analyze data effectively. Many organizations have access to vast quantities of data, but they often lack the sophisticated data mining tools required to make the most of the data. This is a missed opportunity as the data can be used to enhance operational efficiency and increase the police department’s crime-fighting capability. This is where resources need to be highly focused.
Data analysis for counter terrorism
Terrorist incidents are on the rise globally, and while the number of deaths fell slightly in 2021 – dropping 1.2%, according to the Global Terrorism Index – attacks are still largely lethal and highly unpredictable. It is known that terrorists are using increasingly sophisticated technologies, including encrypted messaging systems, GPS and drones.
One of the major challenges in using data to fight terrorism is the need to balance the privacy and freedom of the individual with the need to protect a country’s citizens. The dragnet approach to data gathering is argued by some to be less effective when it comes to terrorism because of the unpredictable nature of many attacks. Identifying an individual who is likely to commit an act of terrorism is frequently likened to finding a needle in a haystack.
Machine learning algorithms are becoming more sophisticated, but there is still work to be done to improve their effectiveness and reliability in the sphere of counter terrorism.
Data analysis for natural disasters
Floods, earthquakes, landslides and droughts have all become more common in recent years, partly due to climate change. The death toll from natural disasters can be high, particularly in low- to middle-income countries, which often lack the proper infrastructure to protect the local population. In higher-income countries, the death toll tends to be lower, partly because there are more disaster risk reduction strategies in place. Globally, around 45,000 people die each year from natural disasters.
Even when the death toll is lower, natural disasters still have other costs, with people made homeless and huge infrastructure repair costs. For example, in September 2022, Hurricane Ian caused damage to infrastructure in the US and Cuba estimated at more than $20 billion.
Advanced technology and predictive data analytics can be used to reduce the worst impacts of some natural disasters. While the number of natural disasters is currently on the rise, the death toll is lower than in previous decades.
The use of disaster detection systems has been a major contributor to the reduction in loss of life. Using data analytics, scientists are sometimes able to give authorities sufficient warning of disasters to allow them to take precautionary measures to minimize the loss of life – for example, through early evacuation of affected areas. When rescue resources are scarce, having data to help prioritize the use of those resources can be invaluable.
Data technologies are now being used by a wide range of professionals who are involved in managing natural disasters, including:
- Specialist computer scientists
- Disaster agencies
They gather and monitor data continuously from sensors, satellites, weather forecasts and even drone footage, using the data to predict weather patterns and alert authorities when there is a need to take action. Information can be used not just to predict an emergency but also the best action to take in response. For example, it can be used to identify the best evacuation routes and rescue strategies when floods or hurricanes are forecast. Officials are also in a better position to plan where to store rescue resources. If they know the typical extent of flooding in a particular area, they can locate rescue equipment beyond the reach of anticipated flooding but close enough to be quickly accessible in an emergency.
Each natural disaster that occurs also contributes to the data available, allowing scientists to gather further insights and have greater accuracy in predicting future incidents.
There are many examples from all over the world of data analytics being used to minimize the harmful impacts of natural disasters.
For example, scientists in Japan have developed a system that analyzes data from earthquake hotspots and issues early warnings of predicted earthquakes to residents in vulnerable areas. This allows them the time to take steps to avoid being caught in earthquakes or tsunamis.
In Australia, predictive analytics now enable flood warnings to be sent to more than 40% of the population when weather patterns suggest that flooding is imminent. This gives residents the opportunity to take precautionary measures to protect themselves and their property against the worst impacts of extreme weather.
In India, the Google Flood Forecasting Initiative has been helping to keep people informed and safe by providing data to governments predicting where and when flooding will occur. Similarly, in Bangladesh, which is affected by flooding more than any other country in the world, Google has been working with the Bangladesh Flood Development Board to deliver flood warnings to citizens to help to keep them safe.
Use of real-time data from survivors
Real-time data from survivors can be used to support rescue operations. This can come from wearables such as smart watches, connected medical devices, and mobile phone apps. First responders can prioritize 911 calls made during an emergency, when there is a spike in call numbers and they cannot respond to all requests for help quickly, based on data such as the age and known medical condition of the callers.
When an emergency has occurred, social media can be used by members of the public in the affected area to provide precise and up-to-date information to other people and officials on local impacts, such as road closures and power outages. People can also use social media to let loved ones know they are safe. Not only is this welcome news to friends and family, but it is also useful information for letting authorities and responders know where to prioritize their resources.
Social media extracts have been used in the past to access information about public safety incidents before they have been officially announced. For example, in 2005, the American Red Cross used social media data to gain insights into Hurricane Katrina,
Location technology is now used in managing many natural disasters. During the Carr fires in Redding, California, officials were able to warn local residents of danger by publishing precise location maps of where the fires were spreading.
Other uses for data science in public safety
There are many other opportunities to use data to enhance public safety. For example, it can be used for identifying patterns and trends in the placement of improvised explosive devices (IEDs) in war zones and former war zones. Data can also be used to inform people and authorities of major road defects in a specific area and to identify accident hotspots.
Data-driven insights could be used in the provision of assisted living services to elderly people. Proactive and reactive safety solutions and notifications can be delivered to customers recommending a course of action to be taken by the customer or a career.
An Italian insurance company has piloted using data to send warnings of extreme weather to its customers so they can take steps to protect their assets. This benefits both the customer and the insurance provider.
Public safety is a rewarding career sector in which you can make a positive difference in people’s lives. Maximizing the potential of big data to support this is an important element of many roles in this sector. Some of the roles that a public safety degree will prepare you for include Chief Security Officer, Police Chief, Crisis Analyst, and Health and Safety Specialist.
In an age where huge volumes of data are being captured, it is essential to ensure that data is appropriately managed and analyzed with the aid of technologies such as artificial intelligence. This enables scientists and agencies to harness the power of data to enhance public safety and deliver real benefits to people all over the world.