Personalised adaptive learning is becoming increasingly popular as a method of providing each student on an online platform with learning experiences that are tailored to their own requirements and peculiarities. This enables learners to go along many learning routes with the shared objective of information and skill development. In such systems, adaptivity and intelligence play distinct roles, with adaptivity being a more data-driven decision-making approach and intelligence being the emulation of human traits in a learning setting. Pedagogical agents, as defined in the field of artificial intelligence, are virtual characters with anthropomorphic features that are introduced for educational reasons. Because e-learning is a continuously growing area, the responsibilities of pedagogical agents change based on the goals that have been established for them. This article provides a systematic evaluation of pedagogical agents’ research and empirical data in e-learning from 2015 to 2022. Their responsibilities will be examined specifically in terms of flexibility and variety, realistic simulation, and their influence on learning: performance improvement, improved motivation, and engagement. The article finishes with a discussion and recommendations on pedagogical agents’ future directions in this ever-changing world of individualised adaptive e-learning.

The main goal of this paper is to provide an overview of the rapidly developing area of software agents serving as a reference point to a large body of literature and to present the key concepts of software agent technology, especially agent languages, tools and platforms. Special attention is paid on significant languages designed and developed in order to support implementation of agent-based systems and their applications in different domains. Afterwards, in the paper a number of useful and practically used tools and platforms that are available and support activities or phases of the process of agent-oriented software development are presented.

Accommodating learning styles in adaptive educational systems represents an important step towards providing individualized instruction. The paper summarizes the main results reported in the literature, as well as the main criticisms of the current approaches. Thus, most of the adaptive educational systems to date are based on a single learning style model and the diagnosis is done explicitly, by means of a psychological questionnaire. The lack of an integrative, comprehensive learning style model and the questionable validity and reliability of the measuring instruments are two important problems of this approach. In this context, we advocate the use of a unified learning style model, which exhibits the following properties: i) it integrates the most relevant characteristics from several models in the literature; ii) it includes e-learning specific aspects (technology related preferences); iii) it is stored as a set of learning characteristics, not as a stereotyping model. Moreover, the student diagnosis is to be done implicitly, based on analyzing her/his interactions with the system.

Last decade, intensive research on emotional intelligence has advanced significantly from its theoretical basis, analytical studies and processing technology to exploratory applications in a wide range of real-life domains. This paper brings new insights in the field of emotional, intelligent software agents. The first part is devoted to an overview of the state-of-the-art in emotional intelligence research with emphasis on emotional agents. A wide range of applications in different areas like modeling emotional agents, aspects of learning in emotional environments, interactive emotional systems and so on are presented. After that we suggest a systematic order of research steps with the idea of proposing an adequate framework for several possible real-life applications of emotional agents. We recognize that it is necessary to apply specific methods for dynamic data analysis in order to identify and discover new knowledge from available emotional information and data sets. The last part of the paper discusses research activities for designing an agent-based architecture, in which agents are capable of reasoning about and displaying some kind of emotions based on emotions detected in human speech, as well as online documents.

Recommender systems have evolved significantly in recent years, using advanced techniques such as explainable artificial intelligence, reinforcement learning, and graph neural networks to enhance both efficiency and transparency. This study presents a novel framework, $X{R}^2{K}^{2}G$ (X for explainability, first R for recommender systems, the second R for reinforcement learning, first K for knowledge graph, the second K stands for knowledge distillation, and G for graph-based techniques), with the goal of developing a next-generation recommender system with a focus on careers empowerment. To optimize recommendations while ensuring sustainability and transparency, the proposed method integrates reinforcement learning with graph-based representations of career trajectories. Additionally, it incorporates knowledge distillation techniques to further refine the model’s performance by transferring knowledge from a larger model to a more efficient one. Our approach employs reinforcement learning algorithms, graph embeddings, and knowledge distillation to enhance recommendations by providing clear and comprehensible explanations for the recommendations. In this work, we discuss the technical foundations of the framework, deployment strategies, and its practical applicability in real-world career scenarios. The effectiveness and interpretability of our approach are demonstrated through experimental results.