Article Preview
Top1. Introduction
Several fields are increasingly moving towards the adoption of ubiquitous technology. The term “ubiquitous computing” was proposed in 1991 by Weiser (1991). This vision describes environments filled with miniaturized computing devices embedded into everyday objects and communications infrastructures. By 2030, future networks will emerge; we will forge links with invisible life and nanoscopic nature (Khalfi & Benslimane, 2013; Lalanda, 2021), see Figure 1.
Figure 1. Evolution of ubiquitous computing (Khalfi & Benslimane, 2013)
The proposed context models are very numerous and diverse. They must ensure efficiently rapid management of contextual COVID-19-Tracer and events, which is rarely the case, (Khalfi & Benslimane, 2015). These models evolve with the technologies offered on the market (Figure 2). Thus, a new vision is needed to minimize these technological dependencies.
This requires the use of engineering approaches and the use of high-level generic models supported by meta-models that facilitate the design of applications and allow the support of all ubiquitous features (Trajano et al., 2020).
We present two theoretical and practical contributions. The theoretical contribution concerns the design of a new intelligent meta model based on the EMF standard by combining it with machine learning.
The second contribution is a Java graphical modeling tool with a drag-drop palette have been provided to allow modeling and visualization of simple and complex ubiquitous scenarios. The proposed graphical model makes it possible to model Android applications, then generate the corresponding code. To ensure the capacity of our development process and the proper functioning of our intelligent framework, we report on the development of an Android mobile application in the field of health. The proposed model allows tracking, contextualizing, and anticipating the risks of COVID-19. To ensure the capacity of our development process and the proper functioning of our intelligent framework, we report on the development of an Android mobile application in the field of health.
The proposed model allows tracking, contextualizing, and anticipating the risks of COVID-19. Noting that our approach is generic and can be applied to any field of application. Classification of COVID-19 patients is offered into two different categories i.e., Infected, and Uninfected based on six different symptoms as input for ANN.
This paper is organized as follows: Section 2 introduces the status of our work. Section 3 details the various contributions related to the subject field resulting in a synthesis. Section 4 details the architectural model of the proposed system. Section 5 presents an instance of the COVID-19 case study MetaModel. Section 6 present discussion of results. Section 7 takes stock of our contributions and presents the research perspectives.