KG-KISS: Knowledge and Information Sharing System with Knowledge Graph

In order to facilitate collaborative decision-making during modern surface warfare situations, locally learned knowledge among sailors and warfighters must be shared effectively in a timely manner. Current Naval approaches for collecting and sharing knowledge are inefficient and inflexible, as new contents are examined over extended timelines with no ability to dynamically update the knowledge base until the next deployment cycle. This effort proposes a peer-to-peer approach to systematically curate and distribute new knowledge within an approved IT infrastructure. The proposed Knowledge Graph Based Knowledge and Information Sharing System (KG-KISS) is based on state-of-the-art heterogeneous knowledge graph representation and machine learning techniques to link knowledge products derived from natural language understanding and content analytics. The goal of the proposed effort is to develop a composable, semi-automated, scalable mechanism capable of dynamically curating multi-media content based on formal rules and machine learning algorithms. The proposed Phase I work will address the leading systems in rapid, automated content curation tools for multiple potential workflows based on the assigned user roles. Realistic use case scenarios will be selected to demonstrate the feasibility of the proposed system with defined operational and technical metrics in a military environment.