The New Eyewitness: Detecting and Solving Crime with Advanced Analytics

by   |   May 8, 2017 5:30 am   |   0 Comments

Scott H. Schlesinger, Chief Analytics Officer, Cognizant Digital Business

Scott H. Schlesinger, Chief Analytics Officer, Cognizant Digital Business

Recently, I watched the movie Patriots Day, marveling at how the Boston Police Department and many other governmental agencies were able to quickly identify and apprehend the Boston Marathon bombers using techniques that weren’t even thought of only a few short years ago. The Boston Marathon bombing stands out as one of the highest-profile example of how law-enforcement agencies have used advanced data and analytics techniques as potent weapons to combat sophisticated and dangerous criminal activities. Today’s law enforcement capabilities contain a mix of advanced analytics, cognitive intelligence and machine-learning tools designed to spot patterns and provide evidence far faster than just a few short years ago. These tools provide indisputable proof of criminal acts in cases where humans either do not witness or are unable or unwilling to provide testimony to solve crimes. The analysis of forensic data, the science of using data to prove activities beyond a reasonable doubt, is at the core of this effort.

Today’s forensic analytics capabilities are becoming the unimpeachable, intimidation-immune eyewitnesses that help solve crimes. Another recent case that is “ripped from the headlines,” highlights the impact of collecting and analyzing data. In Connecticut, police were investigating the murder of a wife whose husband claimed she had been attacked and killed by an intruder in their home while he was out of the house. Police were initially unable to disprove the husband’s events’ timeline because there was little evidence to pinpoint the exact time of her murder. However, police found that the woman was wearing a Fitbit on the morning of the murder. By finding the device and analyzing its data, they proved that she was alive long past when the husband claimed she had been attacked and that he was the only one who could have killed her. In this case, the Fitbit became the virtual—yet indisputable—eyewitness to the crime, providing data that even a few years ago would not have been available (

How, specifically, is this being done?

Advances in data collection, integration, storage, cleansing and analysis have had a significant impact on how effectively guilt or innocence can be determined and a case can be made for punishment, rehabilitation or release of suspects. By leveraging machine learning to mine large and complex data sets, advanced analytics help law enforcement and government agencies connect the dots and find vital information that may be hidden within data from disparate sources, like social media, archived law-enforcement records and third-parties. In some cases, this data has been used to refute seemingly iron-clad alibis of suspects.

Law enforcement is using advanced analytic techniques to detect potentially criminal behavior before that behavior turns into actual crimes. The wealth of available activity and location data, from license-plate readers, to surveillance systems to a variety of financial transactions, gives agencies far better visibility into those activities that are precursors to major crimes.  In some cases, these precursors provide enough information about exact targets so that police can round up the criminals before a crime occurs. In other cases, such as embezzlement or insider trading, advanced analytics can detect the signature of criminal activity long before humans. With the vast quantity of data now available, analytics tools provide the ability to gather, cleanse, store, process, infer and learn at a rate unmatched by today’s resource-stretched agencies. The days where law enforcement can predict who is most likely to commit crimes is coming closer. The “pre-crime” reports depicted in the movie Minority Report is much closer to becoming a reality than a figment of science fiction.

Many of the nation’s top law-enforcement agencies and local jurisdictions are using advanced crime analytics tools to more rapidly process and analyze large volumes of data as part of their criminal investigation process. The do this by sourcing data from mobile-service providers, banking and other financial-services institutions, state and local agencies (departments of motor vehicles and revenue, civil and criminal courts, etc.) and even public utility companies. According to a recent survey done by Wynyard Group, 35 percent of the 300 senior law-enforcement officials surveyed indicated that their agency is currently using some sort of analytics solution to assist with investigations, and 9 out of 10 survey respondents firmly believe that leveraging such tools will become the industry norm near term.

Advanced analytics tools aren’t reserved only for the law-enforcement side of the justice system. While lack of physical evidence or reliable and willing eyewitnesses can result in a criminal not being arrested, prosecuted and incarcerated, inaccurate or incomplete evidence can and has, in some cases, led to the wrongful imprisonment of innocent parties.  Modern tools and processes can also help law enforcement decipher hidden clues not readily visible or recognized by investigators. Artificial intelligence (AI) and machine-learning technology, coupled with business intelligence platforms, can ensure that those guilty of a crime are apprehended while those who may have been caught up (inadvertently) in the justice system and wrongfully convicted (or arrested) are identified and freed from custody.  Organizations such as The Innocence Project use data and analytics to identify and bring forth evidence, especially DNA evidence, to help exonerate the wrongly convicted and reform the criminal justice system to prevent future injustice.

Advances in data identification, collection, integration and analysis have accelerated over the last few years and has impacted nearly every aspect of our lives. New sources and types of data are being acquired and analyzed; data once embedded in email, personal files and unstructured formats are being integrated with traditional/legacy (often structured) data for a more holistic view of the business or specific situation, and data-driven leaders across nearly every sector and geographic region are leveraging new tools and processes, including AI, that improve overall performance.

In the world of law enforcement, rapid use of new, diverse sources of data has in many cases effectively replaced the days of inefficient, manual detective work and turned haphazard guesswork into well-considered and successful discoveries that improve the overall effectiveness of investigative work.  So, the question is, are you living in the past or are you at the forefront, like law enforcement, of harnessing today’s new raw material, data, to its fullest?


Scott H. Schlesinger, Chief Analytics Officer; Cognizant Digital Business

Results oriented technology executive and recognized thought leader with more than two decades of experience and demonstrated success assisting large, global entities in driving organizational change through the leveraging of Information.  Scott is the Advisory leader responsible for the growth, go to market strategy, and overall operations within the business analytics and information management practice globally.


Subscribe to Data Informed for the latest information and news on big data and analytics for the enterprise.



Tags: , , , , , , , , , ,

Post a Comment

Your email is never published nor shared. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>