> Method
The MAISC2 method aims to manipulate the elements of messages within the context of C2 to enrich them to help the recipients understand and interpret them. It has three primary stages: Parameterization, Application; and Presentation. Fig- ure 2 gives an overview of the stages of the MAISC2 method. The Parameterization stage prepares and trains a Machine Learning algorithm to perform entity recognition within the context of C2 systems at hand. This activity is not discussed in detail here, because this work focuses on the trained model’s application and our architectural requirements.
> Researchers
Flavio Mosafi
PhD student in Defense Engineering with an emphasis on Artificial Intelligence at the Military Engineering Institute (IME). Master's degree in Computer Systems at the IME.
Julio Duarte
PhD and master's degree in IT from the Pontifical Catholic University of Rio de Janeiro (PUC-Rio). He is currently a professor in the Postgraduate Program in Systems and Computing at the Military Institute of Engineering.
Maria Claudia
PhD in Systems and Computer Engineering from the Federal University of Rio de Janeiro, Brazil, in 2003. Previously, from 1985 to 2004 he worked as a systems analyst for the Federal University of Rio de Janeiro (UFRJ).
Luis Pires
PhD from Twente University of Technology (1994). He is currently Associate Professor at Twente University of Technology. He has experience in the field of Computer Science.
> Microservice
The MAISC2 microservices architecture, which supports the MAISC2 method, is structured into three layers, namely the Client Layer, the Application Layer, and Pipeline Layer, as illustrated in this figure to the left. Each layer hosts one or more microservices, forming a pipeline encompassing the NLP processing required for enhancing the textual messages exchanged.This enables the Entity Recognition and Priority Classification tasks, as depicted in figure. The following technologies were used to implement these microservices: Docker for creating containers to host the microservices; Django as web development framework; Python, as programming language; and PostgreSQL and SQLite, as the database management systems in the backend.
> Ontology
The ontology proposed in this work is specifically designed to improve the understanding of the textual message exchange within C2 scenarios, while also serving as a basis for record- ing data throughout the entire message exchange application process. It was inspired by conjunct operations of the Armed Forces, with its primary purpose being the identification of key elements that constitute a message exchange in C2, as well as how they relate to each other. Moreover, the elements that result from message processing are also identified. Figure 1 presents a broad view of the complete ontology, where the ontological model is divided into two large blocks, referred to as Block A and Block B, each one representing a component of the C2 message exchange process.
> Taxonomy
