Blue Cross and Blue Shield of Nebraska, a health insurance provider was having a problem with attribution of claims to policy holders. With federal regulations an incorrect correlation would cost them tens of thousands of dollars. The backlog of unmatched information exceeded 2 years. During that time the claims were going unpaid and doctors were leaving the plan.
Contemporary Analysis (CAN) built a solution that looked for matching data points. As the system eliminated obvious matches it continued to connect data points until it paired down the data to a few unknowns. The results were then reviewed by a person to verify or disqualify the results that were inconclusive. Once implemented the system runs in real time and the anomalies are reviewed as part of a regular routine.