Developing Applications for the Internet of Things: Requirements and Platforms
by Everton Cavalcante
Internet of Things (IoT) environments are characterized by a high degree of heterogeneity of devices and network protocols. To deal with such a heterogeneity, several platforms have been proposed aiming at abstracting away the specificities of these devices from applications and end-users, as well as promoting interoperability among them. However, the lack of standardization in IoT makes these platforms adopt different programming models and do not properly address a number of important requirements in this context. Given the relevance of IoT platforms and existing research challenges and opportunities, this talk aims to (i) present the requirements that must be addressed by IoT platforms to allow for the development and execution of applications and (ii) provide an overview of some existing platforms, emphasizing which requirements are met by them.
Everton Cavalcante is an assistant professor at the Department of Informatics and Applied Mathematics (DIMAp) of the Federal University of Rio Grande do Norte (UFRN), Natal, Brazil. He holds both a Ph.D. Degree in Computer Science awarded by UFRN and a Ph.D. Degree in Sciences and Technologies Information and Communication awarded by the Université Bretagne Sud (UBS), France. He has experience in Computer Science with emphasis on distributed systems and software architecture, significantly working on the following topics: middleware, Cloud Computing, Ubiquitous Computing, IoT, smart cities, software dynamic reconfiguration, architecture description languages, and systems-of-systems. He is member of the Software Engineering and Distributed Systems Group at UFRN, where he is the coordinator of the Bachelor of Science in Software Engineering. He is also member of the Association for Computer Machinery (ACM) and the Brazilian Computer Society (SBC), serving as a regional secretary in the latter.
Machine Learning Techniques used in Critical Embedded Systems Applications: What are they? Where do they live? What do they eat?
by Prof. Márjory Da Costa-Abreu
Machine learning is a name given to any technique that can be used in order to find patterns in data. Because of the "Smart-anything" revolution, it is not uncommon to find researchers using machine learning techniques without understanding their meaning and, consequently, not understanding their results. In this talk, we will explore the main problems that can be caused by a bad use of machine learning techniques as well as (hopefully) learn how to avoid such problems using some of the examples from the area of Critical Embedded Systems.
Marjory is a lecturer in Artificial Intelligence at UFRN. She works with Biometrics (psychological and physiological aspects) and identity prediction from a security point of view, Forensic aspects of identity-based data acquisition and processing, Ageing effects in biometrics as well as soft-biometric prediction techniques, Hand-based biometrics (signature, keystroke dynamics, fingerprint, mouse dynamics, hand geometry, touch-screen dynamics), Analysis of medieval handwriting, Data analysis, machine learning and user behaviour in social media. She is part of the editorial board of the IET Biometrics and is a member of the Special Board for Security of the Brazilian Computer Society. She has won the Newton Research Collaboration Programme Award and she has strong collaborations with the University of Kent (UK) and the University of York (UK) and has been working with them in several different projects.