A Sparse Information Processing Refinery for Aviation Sustainment

Oceanit proposes to develop a Sparse Information Processing Refinery (SIPR) targeted to the sustainment of aviation platforms, which is a graphical environment for managing and extracting information from aviation data. Unlike most existing platforms, SIPR is designed for the case where data are sparse and/or difficult to obtain. To make the most of a limited data set, SIPR leverages recent advances in Natural Language Processing (NLP) to ingest handwritten records that provide valuable additional information but are not immediately machine readable. To enhance the processing of that data, SIPR implement a Bayesian inference engine that will enable it to accomplish tasks such as correlating multiple pieces of information or performing root cause analysis and provide feedback about the uncertainty of its predictions. Finally, SIPR leverages the graphical programming environment developed as part of Oceanit’s Versatile Information Processing Architecture (VIPA) to enable a user to rapidly make full use of SIPR’s capabilities.