The concept of analysing increasingly complex data to inform decision making is very relevant to policing today. Data science can and should support modern police forces to serve their communities. But what do we mean by data science and data scientists? One dictionary definition of a data scientist is: “a person employed to analyse and interpret complex digital data, to assist a business in its decision making."

In my first post, I mentioned how the force management statement is changing the way police forces need to think about analytics. Analysts need to expand their analysis from solely looking backwards at what has happened. Now they must also explore what has caused things to happen and develop forecasts and models to understand what is going to happen next. Analysts should be gaining a better understanding of the factors within their data that can predict risk so they can develop tools to minimise the threat.

Data scientists predict and forecast

Police forces can be proactive in the way they use analytics. They can move from reaction to intervention, preventing issues before they arise. Using the same analytical techniques as supermarkets to predict and forecast demand for their products, data scientists can predict and forecast demand for policing resources. Using similar analytical techniques to those adopted by banks to identify risk from data, police forces can process and analyse their data to identify potential threats – before they even happen.

Data scientists will require mathematical and analytical communication and data preparation skills to do their jobs effectively. Within policing or law enforcement they will also need a good understanding of the policing environment and the data that exists within it. Developing a strong understanding of the data and its use will allow data scientists to model the policing context and landscape and develop analytical models to support modern policing.

From theory to empowering

So how can we empower data scientists? I have already talked about the foundation to data science being the data itself. And data scientists will need a variety of data from a wide range of sources at their disposal. But they will also need a variety of tools to enable them to make the most of this data:

• Data preparation tools to quickly and easily prepare data from multiple sources in readiness for analytics.
• Exploration tools to identify relationships and correlations between data items. Data scientists need to test multiple hypotheses quickly and identify the successful models – and “fail fast” with others.
• Modelling tools to develop analytical techniques to predict future trends across policing.
• Visualisation tools to display findings in a variety of ways to customers around the force.
• Publishing tools to automate processes and regularly run models to process and analyse new data.
• Monitoring tools to continually monitor the performance of predictive models. And as older models become less relevant, tools to promote new models into production.

A single platform

Using a single central platform for all these processes will greatly assist the data scientist. It also means that IT governance can support the environment more effectively. With a single place to monitor data lineage and model performance, police forces will gain a greater understanding of what is happening to data throughout the analytical life cycle. And they can track how and why analytical decisions are being made.

Data preparation tools are necessary to allow data scientists to quickly and easily prepare data from multiple sources in readiness for analytics. Click To Tweet

Data scientists will require freedom to explore many different proprietary and open source technologies on a single platform. Empowered data scientists will allow us to unleash the full power of data and analytics.

Click here for more of my thoughts on the role of analytics in policing.