Accurate and relevant knowledge repositories are critical to all organizations and are especially vital to the United States Navy; Sailors' knowledge gaps or inaccurate content could lose billions of dollars and risk human life. Continuous updates to knowledge repositories are a well-documented strategy to avoid the obsolescence of knowledge management systems (Malhotra et al, 2010). In this work, we introduce our instructional content repository, a "YouTube for the Navy" that makes crucial content easily accessible and allows repository maintenance to keep content accurate and up-to-date.
Repository maintenance can be laborious and prone to human error. To increase the accuracy of repository maintenance processes and reduce the need for human labor, our team is applying AI/ML-driven methods to automate and scale repository maintenance for the United States Navy. Leveraging Google Cloud Platform's Vertex AI video intelligence tools and custom algorithms developed by our Carnegie Mellon University partners, we outline several repository maintenance methods that will be tested with a Navy squadron during our Phase II STTR, including Tech Pub Alignment, Comment-Derived Flags, Moment Detection, and Video Segmentation.
Keywords
AI, DATA, ELEARNING, MACHINE LEARNING, PERSONALIZED TRAINING
Additional Keywords
aircraft maintenance