ARES saved the U.S. Government over 960 labor hours through the development of an innovative machine-learning recommendation system. Using MATLAB and Python, ARES data scientists developed an ensemble model to support the binning and categorization of over 15,000 technologies, providing a future-proof solution for classifying these data according to an evolving taxonomy. This has also increased the findability of data, allowing records to be categorized according to a set of classifications to enable structured search mechanisms and aggregate reporting.