Auto and Medical Insurance Fraud

When a hit and run auto collision report uncovers a fraud ring operating out of Los Angeles, investigators need a tool to track and connect knowledge to build a cohesive case for arrest.

Because of the complex nature of this problem, a collaborative and holistic analysis environment is necessary for effective investigation and reporting. Savanna’s shareable workspace and unique, model-based approach are ideal for analyzing complex, inter-agency problems like auto and medical insurance fraud. Savanna’s dynamic all-source analysis environment gives analysts the ability to investigate each point of interest, integrating with existing software, such as SceneDoc’s® mobile reporting application, to form a big picture analysis from all contributing sources.

Savanna Takes on Insurance Fraud

A group of analysts decides that they want to explore a SceneDoc hit-and-run auto collision report that was triggered by a recent crime ring investigation alert.

1: Frame the Problem

First, they use Crumbnet to create a narrative mind map to gain more understanding of the resources needed to prevent insurance fraud. In the Crumbnet, they outline the steps involved in fraud investigations, as well as red flags that could alert investigators to fraudulent activity.

2: Capture Information

With Savanna’s dynamic Occurrence dossiers, the analysts collaboratively populate an knowledge network about a suspected family auto insurance crime ring. Occurrences are building blocks that capture people, organizations, things, places and events related to a problem. In this case, the analysts make an “Edwin Bautista” Person Occurrence. Under its Events section, they add the SceneDoc hit-and-run collision report.

3: Visualize Data to Find Meaning

Linknet
Then, they create a Linknet (Savanna’s link charting tool) to add multiple Occurrences from the knowledge network to visualize connections between Bautista and other family members. They add it to the Insurance Fraud Space (Savanna’s content problem area) for later use.

Timeline
Another way to visualize Occurrence data is with Timeline. In Timeline, the analysts can drop multiple Occurrences, such as various members of the family crime ring Person Occurrences, onto a visual span of time to draw connections between events within each Occurrence. Visualizing event times from multiple Person Occurrences side-by-side provides the opportunity to see similar activity and movement between various crime ring members. To distinguish between events, they customize the colors and date range, bringing the most relevant events front and center.

Map
With the Map tool, they can geospatially visualize a CSV of suspected crimes committed by the Bautista family. The Category lens gives them a detailed view of where each event is located, while the temporal filter shows the analysts a view of the accidents within a certain date range, filtering the data to see the patterns of crime based on location. A screenshot is taken to be used later in a report, which is automatically saved to the Space they’re working in.

Graphic
With Graphic, the analysts can customize Grid (CSV) data to display as bar charts, line charts and more, depending on the data used. Here, they visualize verified staged auto collisions as a bar chart. In the Layers panel, they customize the Graphic by assigning different colors to the categories within a data series, choosing the number of categories they wish to display, or displaying multiple data series on a Graphic at once in order to easily compare information.

4: Discover New Information

At this point, they want to know more about how the original SceneDoc report correlates with the Bautista crime ring, so they’ll use Savanna’s Search tool to find relevant content. Because Savanna’s Search feature can pull indexed mentions of key terms from within PDFs and Analyst’s Notebook Charts®, one of the analysts is able to find a previously built Chart uploaded by another Savanna user outlining the tactics Bautista used to accumulate over $500,000 in fraudulent auto claims.

5: Place Evidence

The analysts revisit the Crumbnet outlining fraud investigations and add discoveries and evidence collected throughout their analysis. From the Space Content panel, they add the Bautista family Linknet and Person Occurrences as supporting evidence to a node. The Crumbnet now acts as an evolving, fully sourced summary of the progressing analysis and is shared with team members and exported to PDF to share with people working outside Savanna.

The Result

Now, with the supporting evidence they have created and gathered, the analysts are ready to compile their findings in a report in Savanna’s Production tool. Productions are reports that help the analysts tell a compelling, fully sourced story about their findings, complete with the content that they’ve gathered and created in Savanna. In their Production, they add relevant content, along with multiple visualizations from their analysis, such as images of the Crumbnet and the Map showing the geo-tagged SceneFiles, as well as a hyperlink to SceneDoc’s app to view similar accident reports. Once complete, the Production is shared directly with team members using Savanna and exported to PDF to send to fellow analysts and decision-makers for further action and prevention.