How Law Enforcement can use Big Data for Safer Communities

New World Public Safety The Call

How Law Enforcement Uses Big Data for Safer Communities

The Call recently sat down with Jeremy Summers, an implementation analyst who trains public safety professionals on how to use intelligence and analytical reporting tools. Summers discussed how these tools help crime analysts, command staff and officers conceptualize the data they collect, store and use to reduce and prevent crime.

The Call: We hear a lot about 'big data' in the public safety software industry. Where does this data come from?
JS: Data comes from everywhere. Think about this: In one police department, you might have 100 sworn officers and 50 civilian staff members. Each of those people deals with hundreds if not thousands of people every year, whether it's through responding to a call for service, taking down a report, making an arrest, booking a suspect or any of the myriad of actions that take place in a police department. Each thing those individuals do creates data, which is collected and stored.

The Call: That seems like it could add up to a lot of data. How is it stored?JS: It's so much data. In some cases, we're talking millions of records. I've worked with agencies that have had more than six million incidents reported over decades. When those same agencies want to get into that information, they need to be able to access it.

That's why, in most cases, public safety agencies store the data in a Microsoft SQL server database so it can be recalled at a later date. This database stores all of the analytical information, which can be anything from names, numbers, addresses, case numbers, anything else that comes into the server, and breaks it down into cubes, so to speak. If you visualize this data in the server, it is structured like a cube with multiple sides full of information. The architecture of the data cubes allows for very detailed information to be extracted and displayed in Microsoft Excel.

Data Cubes

A data cube is a data structure that allows for the fast analysis of vast amounts of information. It manipulates and analyzes data from multiple perspectives. By organizing data in cubes, the limitations of relational databases are eliminated. Users can access and use data for reporting needs in a matter of seconds.

The Call: Talk about these tools and cubes. How do they work?
JS: The capabilities of the right intelligence and analytical reporting tool are just phenomenal. You want a tool that can analyze vast amounts of data and generate data cubes.

The capabilities of this type of tool are really exciting, especially when you consider the vast amount of data that it stores. With a high-level analytical tool that generates data cubes and allows users to take any form of data in their system, analyze it and find correlations, public safety agencies are better equipped to provide the best level of services to the community.

For instance, if you take a data cube and use it to look at data you've pulled from your system, you can identify patterns or trends. A user might notice that accidents are rising in a certain location. With an intelligence and analytical reporting tool, users can see what time of the day the accidents are occurring, what other factors might be coming into play and what can be done to counteract the accidents. This type of tool allows the user to find unknown trends, anomalies or correlations within their data. That doesn't mean the tool itself will say what an agency can do to reduce incidents, but it will help them to see, for instance, that maybe 80 percent of accidents are happening on Main Street while 50 percent of patrol is on First Street.

With this information, resources can be relocated so that accidents or crimes or whatever the issue at hand is can be reduced.

What's really great about this point is that this information can be shown on a heat map. This provides a powerful visual for users to learn from, as it highlights connections that we might not have known existed. You can even drill down into the information to ask it to show all accidents that happened in the past year, the time of day the accidents happened and what the highest number of accidents ever was in that particular location.

It really gives users a better idea of what is happening in the community and helps command staff to better manage their teams because they can see more of what's happening on the streets and where patrol needs to go.

For instance, if users were to look at a heat map to identify burglary trends, they would be able to see within seconds where burglaries were happening, what time of day they happened and when the busiest times of the day or week were for burglaries to occur. It could show, for instance, that most burglaries occurred in a specific area from 6-9 p.m. on Tuesdays. Then command staff could send undercover officers to the area and hopefully catch the individuals responsible for the crime.

A robust intelligence and analytical tool shines a light on patterns and trends so users have actionable information that they can use to make decisions. Ultimately, these decisions could result in lives saved and safer communities.

The Call: What's the difference between a robust intelligence and analytical tool and standard data analysis?
JS: Say an agency needs to know how many tickets were written in a given week. They punch that information into their system and data analysis tools can generate a report right away. It's transactional. It gives you exactly what you asked for, but doesn't give you anything more.

A robust intelligence and analytical tool is going to look at those tickets and instantly slice and dice them and break them down by time of day, day of the week, the frequency in which tickets were issued, what locations in a jurisdiction received more or less tickets, and the percentage of increase for each of those issues.

The thing is, anyone using public safety software spends so much time putting data into their system that they have to have a way to access the information and use it. It's great that they do this and that they can do this, but you have to step back and think about why they're entering so much data. What's the end goal?

The goal is to pull that information out, analyze it and come up with the answers necessary to reduce crime, make response times even faster and provide more safety for the community and first responders.

The Call: To do all of this, it seems like an intelligence and analytical tool must be confusing. How do they work?
JS: I hear that all the time, but in reality, it's a simple tool to use. Essentially, we're able to achieve the immense amount of data analysis because we use Microsoft architecture with SQL servers and Excel.

With Excel pivot tables, we can pull out huge amounts of data in mere seconds. It helps if you picture it like a Rubik's cube; this cube has multiple sides and can contain whatever information you select. Excel has the ability to take that data cube and break it down to show you what you need. It's done in real-time and generates reports in less than 30 seconds. This could take hours with other systems. The analyzing capabilities are truly phenomenal.

The Call: Tell me other ways in which public safety officials use intelligence and analytical tools.
JS: What's really great about solid intelligence and analytical tools is that they are made up of two parts. There's the analytics, which, as I've said, is unlike anything else out there, and then there is the dashboard. Command staff wants to see this information and use it, but they need it to be easily digestible. Enter in dashboards.

Dashboards are a high-level window into the reports generated by the tool. For instance, if you've got someone who doesn't work in the public safety software system every day and isn't that familiar with how law enforcement records management systems work, then dashboards are truly great. They'll show that same person daily updates that will let them look at trends that are occurring at that moment.

A lot of times, command staff just wants to see if burglaries have gone up in a certain area, or if domestic violence calls have gone down. When we know that this is what is needed, we can use the intelligence and analytical tool to generate the answer. Really, all you have to do with dashboards is have the ability to ask the important question you want answered and it will do it for you.

The Call: It sounds like intelligence and analytical tools are really two components. Do these components stand alone or work together?
JS: They're definitely made up of two components. The first of these would be what we call the dashboards. These provide command staff and agency administrators a high-level overview that really works well for grabbing information and showing command staff what they need know. These users need to know when and where officers should go and how many need to be in a specific location for a specific task. Dashboards give that look into crime trends and identify any huge spikes, so command staff has the information necessary to take action.

The second component of a great analytics product is the actual business intelligence tool itself. This portion allows the day-to-day users, such as crime analysts and detectives, to dig deep into the data and start to find trends and correlations that may be hiding within the agencies data.

The Call: Can you summarize the benefits and offer your recommendation for agencies that may be more curious of how data analytics and dashboards may help them?
JS: Of course. These tools aim to provide actionable information, not just data, to help users make better decisions.

What I'd recommended is selecting a system designed around business decisions, not the structure of the data. Look for a system that is super-fast, super-efficient, and allows users from agencies of any size to answer questions they don't even realize they have yet.

The system should offer trend analysis and take huge mountains of data — what seems like insurmountable amounts of data and information — and almost instantly processes it into an easy-to-manage load. A good intelligence program is intuitive and can help users answer questions quickly. If it currently takes hours or days to get a good intelligence report, users are likely to want to look at a system that can complete a comprehensive analysis in a matter of mere seconds. It should generate actionable intelligence to help public safety agencies provide the best level of protection for a community. The system should also be proven and in use by hundreds of law enforcement agencies.

Jeremy Summers is a New World DSS implementation analyst for Tyler Technologies. He has implemented DSS for hundreds of clients throughout the US. Jeremy enjoys empowering crime analysts, command staff and officers with the tools they need to reduce crime and save lives.