Law Enforcement Analytics: Intelligence-Led and Predictive Policing

Fighting Crime With Smart, Timely, and Thorough Data Analysis

For law enforcement professionals, information is critical to fighting crime. Unfortunately, most law enforcement agencies are bogged down by data stored in multiple siloed systems that are difficult to integrate. These organizations are, in effect, data rich but information poor. Law enforcement professionals require intelligent, timely, and complete analysis of data from many sources – including criminal histories, incident reports, crime tips, and 911 calls as well as CAD, RMS, and JMS applications.

Introducing Law Enforcement Analytics

To address these issues, Information Builders has introduced Law Enforcement Analytics (LEA) to make all these data sources available for officers across the enterprise – providing a solution for intelligence-led and predictive policing. LEA combines many technologies – including traditional business intelligence (BI) concepts such as dashboards and scorecards, powerful ad hoc and predictive analytics, interactive mapping capabilities, data mining, and enterprise search. This approach is unique not just in its integration of so many sophisticated technologies, but also in its ability to consolidate, display, and access application data from many source systems in a variety of formats including dashboards, tabular reports, maps, and even in Google-like searches.

With LEA, officers are not forced to connect to multiple systems or master the complexities of separate applications to find answers. LEA presents information in a format that requires little or no training for end users – increasing efficiency and allowing officers to spend less time on redundant clerical tasks.

Dashboards and Scorecards

Police chiefs and command staff can access dashboards containing key performance indicators (KPIs) and other CompStat-type information without having to log in to multiple systems to read data from their underlying systems. This enables them to see what kind of crime has taken place, where and when it occurred, and what resources have been deployed.

Investigators and detectives are presented with a customizable dashboard that provides ad hoc analytics and data mining, enterprise search, and geographical reports, making hard-to-locate data easy to find, and establishing relationships between seemingly unrelated data. Users can set alerts and specify who receives specific types of information. For example, if an investigator is working on a series of robberies, members can be notified – via e-mail or text message, along with a report detailing the incident – as soon as a specific type of call is received.

Finally, patrol officers are presented with dashboards that show all activity and notable events in a given area so they can see what occurred in their patrol areas while they were off duty.

Interactive GIS Mapping

LEA supports geographic information system (GIS) technology by integrating WebFOCUS with ESRI's ArcGIS Server suite to allow bidirectional reporting and mapping capabilities. A user can view report results displayed on an interactive map, as well as in detailed or summarized listings. If a map displays a variety of incidents, users can zoom to a smaller section to get a report. The map and the report can be displayed on the same page to help users with their analyses. This is available from a Web browser and requires minimal training.

Enterprise Search

Enterprise search provides crime analysts and investigators with the ability to search across criminal justice information systems without knowing exactly where the information exists. This enables investigators to find a needle of information in the haystack of data systems. Users can perform simple keyword searches across structured and unstructured data – such as free-form text fields in incident reports – to locate information quickly, and to spot relationships they never knew existed.

Predictive Analytics

Processing large amounts of data and assessing the relative importance of each variable takes time and requires special skills. Operational users do not have the time, know-how, or functional responsibility to analyze data in this way. They need the ability to make decisions quickly, and that is where automation comes into play.

LEA can do in seconds what typically takes analysts days to complete. It provides the ability to run statistical models and scoring algorithms to make predictions about the future to enhance planning and decision-making.

If an agency wants to effectively allocate patrol efforts, it can perform sophisticated analysis with a large number of variables for several years of incident data. Using traditional solutions, this would be a daunting endeavor. The data set would be too large and complex to place into a spreadsheet, and the vague but statistically significant relationships would be difficult to discern. Using LEA, however, users can simply select a few parameters from a Web form, and the statistical model predicts the probability of crimes occurring within each dispatch zone. The results are displayed on a map listing dispatch zones and highlighting the highest crime probabilities.

Social Media Integration and Analytics

Whether its finding a missing puzzle piece to crack a difficult case, identifying new threats, identifying criminal associates, monitoring communications to head-off flash mobs, or uncovering incriminating evidence, LEA enables investigators to perform Google-like searches across popular social media platforms, such as Facebook and Twitter. With social media adaptors and Magnify, investigators can run ad hoc reports and analyze real-time information from tweets, posts, comments, likes, shares, and recommendations. In addition, with Word and Sentiment Analytics, advanced algorithms break sentences into component parts, remove common words, and provide visualizations of the most meaningful words and the frequency of their use over time.