Kunyoung Kim, Donggyu Kim, Hwiseung Son, Mye Sohn, Journal of Internet Computing and Services, Vol. 26, No. 1, pp. 193-202, Feb. 2025
10.7472/jksii.2025.26.1.193, Full Text:
Keywords: Command and control, AI model, Integrated DB, Ontology, Knowledge Graph

Abstract

To train AI models for Joint All Domain Command and Control (JADC2), it is essential to secure training data. To this end, the military is promoting the construction of an integrated DB in the combat cloud. However, building an integrated DB from scratch is a difficult task that requires a lot of time and effort. To reduce these difficulties, this paper proposes a method to build an integrated database using data stored and managed in the command and control (C2) systems operated by the Joint Chiefs of Staff (JCS) and each military branch by utilizing ontology and knowledge graphs. At this time, ontologies are used to integrate structured data stored in the JCS and each military branch's C2 DB, and knowledge graph is used to integrate unstructured data such as documents. The ontology structure consists of several domain ontologies and an upper ontology that can integrate them, considering that there are many systems related to C2 and the related areas are broad. Afterwards, the illustrative scenario is constructed to show it is possible to effectively integrate and obtain data from multiple sources that can be used for learning AI models based on the proposed method.

Statistics
Cite this article