Analysis of Security of Machine Learning and a proposition of assessment pattern to deal with adversarial attacks

Today, Machine Learning is being rolled out in a variety of areas. It is a promising field that can offer several assets and can revolutionize several aspects of technology. Nevertheless, despite the advantages of machine learning technologies, learning algorithms can be exploited by attackers to carry out illicit activities. Therefore, the field of security of machine learning is deriving attention in these times so as to meet this challenge and develop secure learning models. In this paper, we overview a taxonomy that will help us understand and analyze the security of machine learning models. In the next sections, we conduct a comparative study of most widespread adversarial attacks then, we analyze common methods that were advanced to protect systems built on Machine learning models from adversaries. Finally, we discuss a proposition of a pattern designed to ensure a security assessment of machine learning models.

Secure Learning Management System Based on User Behavior

The COVID-19 outbreak is an international problem and has affected people and students all over the world. When lockdowns were imposed internationally, learning management systems began to be used more than in the previous period. These systems have been used also for traditional forms of learning and not only for online learning. This pandemic has highlighted the need for online learning systems in the educational environment, but it is very important for these systems to be secure and to verify the authenticity of the students when they access a course or evaluation questions. In this period, everything is moving towards the digital world, with students that are connected from a distance to online systems. All activities in the educational environment will soon be performed digitally on learning management systems, which includes also the evaluation process of the students. In this paper, we propose a secure learning management system that uses the student’s behavior to identify if they are an authentic student or not. This system can support the teacher’s activities in the learning process and verify the authenticity of the students logged on to the system. This paper is aimed at learning management system developers, who can use the proposed algorithms in their developed platforms, and also at teachers, who should understand the importance of the identification of students on these platforms.

What do young adults’ educational experiences tell us about Early School Leaving processes?

What leads young people to disengage and leave school early? This paper focuses on young adults’ educational trajectories, with the objective of identifying useful insights for improving school measures and strategies against early school leaving and school disengagement. Specifically, it draws upon an empirical study with young adults from the north of Portugal. Longitudinal bio-interviews took place with young men and women, aged 17 to 24, who were enrolled in secondary education at mainstream schools or in alternative learning contexts or who were early school leavers. A set of critical moments and circumstances were identified as turning points in the young adults’ descriptions of their school trajectories, revealing both obstacles and incentives to continue studying. Based on these narratives, the study identifies some risk and protective factors for early school leaving and makes suggestions about how to improve school-based measures to prevent school disengagement. Some key recommendations for schools include: increased support mechanisms during educational transitions; a firm stance on bullying and the creation of violence-free and secure learning environments; enhancing socio-emotional support and encouraging the emergence of relationships of care; and providing meaningful education by ensuring the school content is aligned with students’ motivations and needs.

Maintaining Quality in Online Learning Environments

The online classroom continues to play an ever increasing role in higher education. There are proven, research-based pedagogical techniques available to instructors who want to create online courses that are both dynamic and engaging. With careful planning, online learning can provide students with a positive learning experience without sacrificing the academic quality of learning. Addressing security issues and challenges is vital to maintaining the desired academic rigor and quality. This paper discusses these important security issues and offers a variety of solutions for facilitating a secure learning environment.