Accurately classifying user inputs into pedagogical ontologies is essential for conversational Artificial Intelligence (AI) systems in educational technology. The MEPA ontology comprises seven specialised pedagogical concepts designed to represent personal learning experiences, and presents unique classification challenges due to its domain-specific nature and conceptual nuances. This paper evaluates the ability of state-of-the-art large language models to perform this specialised classification task effectively. Multiple LLM variants were tested, including GPT-5 (Standard, Mini, and Nano) and the Claude model families. All were configured with extended reasoning capabilities and optimised prompts. Despite these optimisations, LLMs achieved suboptimal classification performance on our test dataset of 350 expert-labelled samples. In response, we developed a specialised classification system that combines OpenAI’s Text Embedding 3-Large model with a feedforward neural network that was trained on 14,000 balanced samples. Our custom model achieved 94% accuracy, outperforming the best LLM baseline by 12%. We analyse the factors contributing to the limitations of LLMs in highly specialised ontological contexts and discuss the practical implications for educational technology systems that require precise alignment with pedagogical frameworks. Our findings suggest that domain-specific models can significantly outperform general-purpose LLMs for specialised classification tasks.

In the current digital era, smartphones have evolved beyond simple communication tools into essential multifunctional devices that particularly pervade the daily lives of young adults. This article systematically reviews the impact of smartphone usage on the digital wellbeing of university students. Through a structured review of empirical studies, we examine the relationship between smartphone use and various aspects of digital wellbeing, such as sleep deprivation, attention deficits, and social challenges. We discuss the methodologies employed in these studies, analyze the data collected, and summarize and categorize the findings within the current research. This comprehensive synthesis aims to inform scholars and policymakers of the impact of smartphone usage on digital wellbieng and suggests potential directions for future research into digital wellbeing.

Social media have become an indispensable part of daily life, particularly among university students, who regularly browse social news feeds in their spare time. Due to their pervasiveness, social media platforms provide an opportunity for influencing user behavior and encouraging informal learning. In this paper, we present an experiment using an online video recommendation application designed to blend micro-informative content with general content according to user preferences and activity history. Based on a one-week study, we conclude that injecting micro-informative content into video streaming platforms has the potential to improve the perceived satisfaction of users and can act as a potential catalyst to motivate users to consume more informative content online.

Over the past decade, the use of chatbots for educational purposes has gained considerable traction. A similar trend has been observed in social coding platforms, where automated agents support software developers with tasks such as performing code reviews. While incorporating code reviews and social coding platforms into software engineering education has been found to be beneficial, challenges such as steep learning curves and privacy considerations are barriers to their adoption. Furthermore, no study has addressed the role chatbots play in supporting code reviews as a pedagogical tool. To help address this gap, we developed an online learning application that simulates the code review features available on social coding platforms and allows instructors to interact with students using chatbot identities. We then embedded this application within a lesson on software engineering best practices and conducted a controlled in-class experiment. This experiment examined the effect that explaining content via chatbot identities had on three aspects: (i) students’ perceived usability of the lesson, (ii) their engagement with the code review process, and (iii) their learning gains. While our findings show that it is feasible to simulate the code review process within an online learning platform and achieve good usability, our quantitative analysis did not yield significant differences across treatment conditions for any of the aspects considered. Nevertheless, our qualitative results suggest that students expect explicit feedback when performing this type of exercise and could thus benefit from automated replies provided by an interactive chatbot. We propose to build on our current findings to further explore this line of research in future work.

Peer code review has proven to be a valuable tool in software engineering. However, integrating code reviews into educational contexts is particularly challenging due to the complexity of both the process and popular code review tools. We propose to address this challenge by designing a code review application (CRA) aimed at teaching the code review process directly within existing online learning platforms. Using the CRA, instructors can scaffold online lessons that introduce the code review process to students through code snippets, following a format resembling computational notebooks. We refer to this online lesson format as the code review notebook format. Through a case study comprising an online lesson on code quality standards completed by 23 university students, we evaluated the usability of the CRA and the code review notebook format, obtaining positive results for both. These results are a first step toward integrating code review notebooks into software engineering education.

Chatbots have long been advocated for computer-assisted language learning systems to support learners with conversational practice. A particular challenge in such systems is explaining mistakes stemming from ambiguous grammatical constructs. Misplaced modifiers, for instance, do not make sentences ungrammatical, but introduce ambiguity through the misplacement of an adverb or prepositional phrase. In certain cases, the ambiguity gives rise to humor, which can serve to illustrate the mistake itself. We conducted an online experiment with 400 native English speakers to explore the use of a chatbot to harness such humor. In an interaction resembling an advanced grammar exercise, the chatbot presented participants with a phrase containing a misplaced modifier, explained the ambiguity in the phrase, acknowledged (or ignored) the humor that the ambiguity gave rise to, and suggested a correction. Participants then completed a questionnaire, rating the chatbot with respect to ten traits. A quantitative analysis showed a significant increase in how participants rated the chatbot’s personality, humor, and friendliness when it acknowledged the humor arising from the misplaced modifier. This effect was observed whether the acknowledgment was conveyed using verbal, nonverbal (emoji), or mixed cues.

Spending an uncontrolled quantity and quality of time on digital news and social media platforms can negatively influence mental health and decrease cognitive abilities. In this paper, we propose a sequential news recommendation system employing deep reinforcement learning to capture the user’s short and long-term interests while blending social news with micro-learning informative news items that can help users derive useful outcomes out of their online presence. In the absence of a publicly available dataset, we developed a simulation model to synthesize data and evaluate the proposed news recommendation system. We train and evaluate our model on synthesized data and show an improvement in user satisfaction.

This paper presents the MUST-VIS system for the MediaMixer/VideoLectures.NET Temporal Segmentation and Annotation Grand Challenge. The system allows users to visualize a lecture as a series of segments represented by keyword clouds, with relations to other similar lectures and segments. Segmentation is performed using a multi-factor algorithm which takes advantage of the audio (through automatic speech recognition and word-based segmentation) and video (through the detection of actions such as writing on the blackboard). The similarity across segments and lectures is computed using a content-based recommendation algorithm. Overall, the graph-based representation of segment similarity appears to be a promising and cost-effective approach to navigating lecture databases.

This paper reports on the successful use of Graasp, a social media platform, by university students for their collaborative work. Graasp features a number of innovations, such as administrator-free creation of collaborative spaces, a context-aware recommendation and privacy management. In the context of a EU-funded project involving large test beds, we have been able to extend this platform with lightweight tools (widgets) aimed for learning and competence development and to validate its usefulness in a collaborative learning context.

This paper discusses the adoption of bottom- up social software tools in formal learning environments. This is believed to enhance the learning experience of today's young generation characterized by being technology savvy and keen on social networking. As a first step towards this objective, the 3A interaction model that aims at aiding the design of personal and collaborative learning platforms is presented. It accounts for interaction paradigms widely used in Web 2.0 applications and builds on Distributed Cognition and Activity Theory while remaining at the right level of abstraction to be easily ¿translatable¿ into tangible applications supporting both formal and informal learning.

This paper presents the findings of a study on the acceptability in higher education of a Web 2.0 collaborative application, namely eLogbook. The latter offers several features for sustaining collaboration and supporting personal and group learning. It was introduced to students taking a laboratory course that spans over one semester and mainly consists of in-class group experiments. In this paper, we present eLogbook. We then describe our hypotheses as well as the qualitative and quantitative methods used to evaluate the usefulness and usability of eLogbook in a formal learning context. Finally, we discuss our findings and its implications.

In this paper, we propose a personalized and contextual ranking algorithm implemented on top of the 3A interaction model. The latter is a generic model intended for designing and describing social and collaborative learning platforms integrating Actors, Assets and group Activities (the 3 "A"). The target user's interactions with his/her environment are modeled in a heterogeneous graph. Then, the algorithm is applied to simultaneously rank actors, assets and group activities taking into account the target user and his/her context. As an illustrative application and a preliminary evaluation, we apply the algorithm on data related to the activities carried out in a European Research Project, especially the collaboration between its members through the joint production of deliverables in workpackages.

In the framework of the European Integrated Project PALETTE, the Ecole Polytechnique Federale de Lausanne (EPFL) is developing the eLogbook Web 2.0 social software. The purpose of eLogbook is to support tacit and explicit knowledge management in communities of practice. It can be customized by the users to serve as an asset management system, as a task management system or as a discussion platform. In this paper, the innovative Computer-Human Interaction features of eLogbook are introduced and its deployment scenario to support collaborative laboratory activities in engineering education is described. The main idea is to sustain interaction for learning purpose within self-organized teams that integrate -on a seamless level- both human actors (students, teaching assistants) and non-human actors such as laboratory equipments or software agents.