An IoT-based mobile gateway for intelligent personal assistants on mobile health environments

Using Accuracy Analysis to Find the Best Classifier for Intelligent Personal Assistants

The wide dissemination of Internet around the world changed the way people communicate among them. The evolution of the information and communication technologies (ICT) or artificial intelligence (AI) allowed the creation of new types of services involving electronic devices with huge potential. The emergence of intelligent personal assistants (IPAs) is one of such potential example. Combining the IPA concept with the recent paradigm of Internet of Things (IoT), offers new possibilities and new services to end users, since they could learn more about other entities present in the surrounding environment. This paper proposes a novel mobile gateway solution for a ubiquitous mobile health scenario, in which information gathered from a body sensor network (BSN) is used by an IPA belonging to another person, typically mentioned to as caretaker. The mobile gateway receives real time information related to location, heart rate, and possible falls of the monitored person and acts as a communication channel between the BSN sensors and the IPA platform. Furthermore, the paper presents a performance evaluation study of the proposed mobile IoT-based gateway demonstrating and validating its feasibility.

Internet of things mobile gateway services for intelligent personal assistants

Virtual assistants are becoming more common today as access to this technology has become cheaper and more affordable to the common consumer. As a result of this increase in demand, a component based software engineering approach may be more practical in the future development of these systems. This article presents a study of the current state of the research on component based software engineering and service principles in virtual assistants. The main motivation for this research is that there just is not any comparative study on how component-driven development might benefit our virtual assistants, especially those with voice interaction. This study is limited to papers that go into detail about the development of intelligent personal assistants and the entities that comprise them. Due to the nature of virtual assistants, we initially assumed that many of them would primarily be made of various software components. However, our results show that the development of virtual assistants is actually more complex than we initially believed, and that a more consolidated approach is necessary to define the best method of development for virtual assistants.

/pdf/component-driven-development-in-modern-virtual-assistants-a-au9qgkg0ta.pdf

Component-Driven Development in Modern Virtual Assistants: A Mapping Study

A considerable number of studies have been made on ambient service management platform for smart home environments. Few of them addressed the robustness issue, which is fundamental for such environments. Robustness is critical because there are usually services that are critical to the habitants health and life in Smart Homes. In this paper, we propose an efficient and adaptive failure detection and recovery protocol for an ambient service management platform, in which failures are detected efficiently with a novel rotating roll-call algorithm without a centralized monitor. Adaptive techniques are also proposed to maintain network stability. The analysis and experiment results show that the proposed solution is capable of detecting software failures efficiently without causing network instability, hence preserving robustness of an ambient service platform at home.

Carlos Romeiro

Papers

A Pervasive System Architecture for Smart Environments in Internet of Things Context