Creating Mission Focused AI Solutions
Intelligence
Customer Success Stories
26% Data Quality Improvement
Information from different sources found in most data warehouses contain conflicts that are hard to find and resolve. Differentiating between similar people, places, and organizations is essential to accurate and actionable intelligence. Common approaches for deconflicting data use static rules for choosing an authoritative representation. Our novel use of knowledge graph relationships allows us to develop ML models that identify and correct erroneous, conflicting, and deficient information. In a recent project data quality increased by 26% based on federal standards.
Using Analytics to Reveal Terrorist Networks
JITF-CT was a cross COCOM (USNORTHCOM, USCENTCOM, USSOUTHCOM, USEUCOM, USAFRICOM, USSOCOM) tool to identify then Find, Fix, and Finish terrorists. Using intelligence reports, analysts created link charts of terrorist networks to associate different entities together, but the volume of data was overwhelming the analysts causing lost opportunities and wasted time. After Analyst Shadowing sessions and Day in the Life (DITL) analysis, we discovered that the link charting tools were cumbersome and unintuitive, and the networks were so big that human analysis needed to be augmented with machine analytics. To address these issues, we built an application that enhanced the COTS software by making it as easy as possible to create multiple links in fewer steps. Simplifying the linking process revealed previously unknown links between people of interest. Link creation rates improved by 400%, and the program was credited with numerous captures and kills.
Automation to Enable Low-Cost Cloud Migration
Legacy systems are present in most agencies but can carry significant security risks. Building a modern cloud-based system and maintaining continuous operations while transitioning legacy systems is essential for mission critical systems. SphereOI uses process automation to stabilize legacy systems. For a recent program, we increased legacy system availability from 96% to 99.8% and decreased the team from 30 FTE to 8. The remaining FTEs were applied to new system development in parallel with legacy stabilization. As part of any system update, SphereOI conducts outreach programs to inform system users of changes to interfaces. In the Intelligence Community, changes must be coordinated well in advance to ensure critical systems are updated before any deprecation. This outreach enables legacy systems to be retired gracefully rather than running past EOL and risking failure or never being retired due to unresolved dependencies. In a recent program, we retired the legacy system, migrated more than 13TB of data, and transitioned more than 400 client systems to a new system with no downtime and no increase of funding.
