Self-Healing Data (ADCAT)

Data is one of the Department of the Navy (DON)’s greatest assets, providing vital information to numerous functional communities. Poor data quality due to human-generated errors and errors injected when data flows between information systems is hindering DON’s ability to gain valuable insight from their data. Illumination Works (ILW) offers DON the Automated Data Cleansing and Analysis Tool (ADCAT), a robust, platform-agnostic, modular solution for data cleansing that can be applied at the point of data entry or on the backend. By leveraging outlier detection, machine learning, and probabilistic graphical modeling techniques, ADCAT identifies and heals errors in a time-efficient and automated way. During Phase II, ILW will enhance and harden ADCAT for DON operational use to achieve improved data quality across functional areas and data types. Phase II Base efforts will expand ADCAT to accommodate additional input datatypes and enhanced to incorporate an explanation module that provides users with insight as to why a data correction is suggested. During Phase II Option, ILW will demonstrate ADCAT’s ability to interface with a DON operational environment and heal data errors therein. ILW will enhance ADCAT to an agent architecture that seamlessly integrates inbound and outbound routing with ADCAT’s core cleansing components.