Adapterless Information Consolidation

Sharing data between software systems developed at different times and with different functional goals has proven notoriously difficult. Today’s disparate software systems are still largely integrated via manual development of custom “adapter” software. Adapters increase integration and lifecycle costs, sometimes cause significant delays in useful access to source data, and introduce errors. The costs of adapter-based solutions are particularly acute for today’s Army due to increased costs, deployment delays, and the resulting impacts on tactical operations and missions. An “adapterless” approach to information sharing and consolidation would mitigate the costs and provide substantial benefit to the Army. Consolidation via Ontology-Driven Extraction, Semantic Mapping, and Adaptation for Real-World Translation (CODESMART) will provide an adaptable, scalable software component that collects information from arbitrary sources and translates it into common, readily accessible forms. Actionably-accurate semantic representations of the data within an application are created by using emerging methods and tools that formally model application domains and the software applications themselves. Recent advances in machine learning and agent reasoning provide a technical foundation to extend prior model-based extraction to make it sufficiently adaptable for fieldable use.