Data can appear lifeless and dull on the surface – especially government data – but the thought of it should actually get you excited. Data is a very interesting and powerful thing. First off, data is exactly the stuff we bother to write down – and for good reason. But its potential far transcends functions such as tracking and bookkeeping: Data encodes great quantities of experience, and computers can learn from that experience to make everything work better.
For example, take agriculture – and the federal studies that advance it. A farmer jots down crop rotations as needed to manage operations. And later this data also serves to track crop productivity. But for the real payoff, the farm boosts productivity rather than only tracking it. By number-crunching its records, it discovers ways to optimize operations: which crop schedule, seeds, treatments, irrigation options, fertilizers and other process decisions best increase crop yield. By learning from the data, the farmer notes that notes are more notable than previously noted.
The need for optimization is palpable. By the end of the century, agriculture will have to deal with a population of 11 digits – a vast number of mouths to feed – in combination with a diminishing supply of available land. This era of exploding magnitudes decidedly demands data science. Yet the growth in scale is actually good news for predictive analytics. It presents a greater opportunity than ever, for two reasons. First, it means more “fuel for intelligence,” i.e., more data from which to learn. And second, larger-scale operations themselves stand to benefit that much more when optimized – the returns are commensurate.
This principle applies far beyond agriculture, manufacturing and even the industrial sector as a whole. Data science drives growth and efficiency across verticals, including financial services, insurance, retail, e-commerce, energy and healthcare, while bolstering business functions such as sales, marketing, advertising, customer service, human resources, risk management and supply chain management.
In government, data science’s vital impact extends just as far and wide, empowering agencies to more effectively serve and safeguard citizen fundamentals such as health, safety, housing, economic stability, education, equality and justice. Here are some more specific areas where predictive analytics bolsters the effectiveness of government:
Health and safety: Government agencies target what buildings, restaurants and manholes to inspect by predictively modeling which have the greatest risk of fire, health code violations, lead paint poisoning or other safety incidents. The EPA plans to use predictive analytics to regulate air emissions, monitor water quality and avert environmental catastrophes. And the CDC applies predictive modeling to improve population health.
Law enforcement: As is well known, police analytically predict crime, and judges and parole boards pay heed to recidivism risk scores. But a range of other agencies also employ data science to enforce laws and regulations. New York City analytically flags for possible illegal apartments, misused business licenses and other ducked regulations. And the pertinent departments predictively target fraud auditing of tax returns, government invoices, government contracts, workers’ comp, and Medicaid and Medicare claims. Florida’s Department of Juvenile Justice determines rehabilitation assignments based on the predictions of future repeat offenses. And other analytical efforts target internal investigations of potential police misconduct and other forms of injustice.
Defense and homeland security: Military agencies analytically predict threats and civil unrest, while the NSA and FBI predict terrorism. Detecting possible hacker or virus footprints toughens cyber security. The U.S. Department of Defense applies data science to target critical internal operations such as special forces recruitment (predictive hiring decisions), proactive veteran suicide intervention, and the maintenance of those Army vehicles at a higher risk of impending failure.
Predictive analytics is the Information Age’s latest evolutionary step. We have moved beyond engineering infrastructure that stores and manages big data to implementing science that makes actionable use of the data, tapping its contents to optimize most every large-scale activity. The breadth of examples listed above signals that predictive analytics’ role is well established, a status that is further upheld by the many other application areas we see covered at the Predictive Analytics World for Government conference (an offshoot of the PAW event series).
But the fortification of government with data science has only just begun. The pressure mounts as citizen needs intensify, international competition escalates, and infrastructure and security risks grow. Critical measures for alleviating these pressures include reducing waste and abuse, increasing the effectiveness of both targeting and triaging, and optimizing operations for efficiency. Data-driven optimization is a key method for achieving these improvements.
To more fully broaden the role of data science in government – and thereby seize the tremendous opportunity of today’s data eruption – agencies must collaborate. Given the complexity of both managing and analyzing big data, the “use your data!” rallying cry must not only mobilize analytics internally, but also call for sharing technological resources and best practices, for coordinating efforts and for investing in data interoperability. If a small farm has limited data, it relies on cooperation that pulls together data across many farms.
To aptly serve the needs of citizens, government agencies must advance and expand the deployment of data science. If you’ll allow a mixing of metaphors, you can bet the farm on tools that harvest insights from data and cultivate prosperity. And as you take your next steps in contributing to this historic development, the book for which this is the foreword, “Federal Data Science: Transforming Government and Agricultural Policy Using Artificial Intelligence,” guides the way; it has been ideally crafted for that very purpose by a select, international group of experts who come from a diverse range of government and industry backgrounds.
Eric Siegel, Ph.D., is the founder of the Predictive Analytics World conference series –which includes PAW Government – and executive editor of The Predictive Analytics Times. He is the author of the award-winning book, “Predictive Analytics: The Power to Predict Who Will Click, Buy, Lie, or Die,” a former Columbia University professor and a renowned speaker, educator and leader in the field.
This article is excerpted from Siegel’s foreword to the recently released book, “Federal Data Science: Transforming Government and Agricultural Policy Using Artificial Intelligence,” edited by Feras A. Batarseh and Ruixin Yang. Republished with permission.