An international team of mathematicians, led by Lehigh statistician Taeho Kim, has introduced an innovative method that could significantly improve how scientists make predictions, especially in fields like health, biology, and the social sciences.

The results have been impressive by the metrics that matter to technology teams: accuracy rates, processing speeds, cost reduction, throughput. A fraud model that catches 94% of suspicious transactions where a rules engine caught 71% is, by any reasonable measure, an improvement.