This study explores the multifaceted attitudes of high school students toward the use of artificial intelligence (AI) and large language models (LLMs) like ChatGPT in educational contexts. Drawing upon a tripartite model of attitudes, our research evaluates affective, cognitive, and behavioral dimensions to offer a nuanced understanding of students’ perceptions. The affective dimension assesses emotional responses to AI tools, the cognitive dimension examines beliefs about the utility and ethical considerations of AI, and the behavioral dimension evaluates actual usage patterns of AI technologies. Utilizing a newly developed survey instrument tailored for the educational context, data was collected from 93 high school students across different regions of Italy in the period that ranged from February 2024–March 2024. Exploratory factor analysis (EFA) was employed to explore the underlying structure of the survey instrument and identify underlying factors influencing AI acceptance. The analysis reveals three distinct factors—Mindful AI Learning, Embracing AI Effects, and LLM as Learning Companion, highlighting the complexity of students’ attitudes toward AI. Results indicate a cautious but optimistic reception of AI in education, offering crucial insights into Information Intelligence for enhanced learning and the design of personalized learning pathways. The study contributes to the literature by offering a novel scale to measure attitudes toward artificial intelligence, specifically focusing on both general AI and Generative AI large language models, such as ChatGPT. Moreover, it highlights the critical need for AI literacy, ethical digital learning frameworks, and robust institutional policies to bridge the digital divide. Consequently, this work is framed as a preliminary exploratory investigation. Ultimately, these findings advance our knowledge of transformative digital learning processes and inform future strategies for human–machine integration in educational systems.

Social media conversations often face threats such as toxicity, hate speech, fake news, and moral outrage. In particular, moral outrage contagion on social media has harmful consequences, such as the spread of conspiracy theories, protests, and polarization. Teenagers, being intensive users of social media platforms, are more exposed to these threats. Recently, AI and human collaboration is proven to be effective in different applications. Hence, the importance of detecting morality in online posts and effectiveness of human–AI collaboration inspired us to conduct a pilot study with teenagers to identify their behavior with regard to AI suggestions. In our study, we first examine several AI-based methods to identify the moral content from social media platform such as X (Twitter), using a human-annotated gold-standard Moral Twitter Corpus. We then design a multi-stage pilot study with secondary school students to examine how students interact with AI in moral classification, analyzing their responses, decision shifts, and influence of external factors such as social media usage and gender. Our findings reveal both benefits and risks of AI in ethical decision-making. While AI enhances moral content detection and influences decision shifts, particularly from non-moral to moral, it does not fully dictate students’ choices. AI explanations provide some benefits, but their impact remains inconsistent, and AI predictions that diverge from gold-standard annotated labels can mislead students when accepted uncritically. Although social media usage and gender shape moral reasoning, no statistically significant interaction with AI influence on moral decision-making was observed within the scope of this pilot study. These insights emphasize the need for responsible AI, ensuring that students engage critically with AI-generated moral judgments rather than relying on them blindly.

Bloom’s Taxonomy plays a central role in assessment design by helping instructors align evaluation tasks with learning objectives. However, applying Bloom’s framework in practice, especially in programming education, requires substantial effort and often leads to divergent interpretations among educators. This study explores the extent to which Large Language Models (LLMs) can support the automated classification of programming assessment items across Bloom’s cognitive levels. We evaluate Bloom-based classification on a dataset of items from an introductory undergraduate Computer Science course, covering four cognitive levels (Remember, Understand, Apply, and Analyze), comparing proprietary LLMs with open-source alternatives. Our methodology considers two strategies: zero-shot prompting and a council-based ensemble approach. Results show that proprietary models achieve accuracies of up to 78 https://doi.org/10.5281/zenodo.19332485 .

Rapid progress in natural language processing (NLP) has ushered in a new era of artificial intelligence (AI) models. Large language models (LLMs), powered by deep learning algorithms, excel in understanding and generating coherent human language responses, revolutionizing various fields. In education, these models promise to enrich the learning experience by providing personalized support to students and facilitating communication between students and teachers. In particular, Generative Pre-trained Transformer (GPT) technology provides instant and contextually relevant responses, and it has become an increasingly interesting topic in education. This article explores the potential of ChatGPT to enhance learning activities in secondary schools, with a specific focus on concept maps, a valuable tool for fostering meaningful learning. Concept maps are widely recognized for their educational benefits but existing tools for automatic concept map generation lack flexibility and semantic understanding. In contrast, ChatGPT's adaptability and responsiveness make it a compelling candidate for generating concept maps tailored to individual needs. This paper sheds light on ChatGPT's potential in education and offers insights into its effectiveness in supporting learning activities through concept map generation. We present a study involving 83 secondary school students to test their perceptions of ChatGPT-generated concept maps. We compared concept maps created by both ChatGPT and teachers across six different topics. Using PlantUML, we standardized the concept maps, and then used various questionnaires to assess their quality, effectiveness, and impact on student performance. The statistical analysis revealed that the concept maps generated by ChatGPT were comparable in quality to those produced by teachers. This highlights the usefulness of ChatGPT in supporting structured learning activities while also significantly reducing the time and effort required to create concept maps.

This study explores the use of explainable artificial intelligence in education, with a focus on its relevance for Learning Analytics. The research introduces a prototype-based dynamic incremental classification algorithm, Dynamic Incremental Semi-Supervised Fuzzy C-Means (DISSFCM), which leverages fuzzy logic to analyze student interaction data from virtual learning platforms, even when the data are only partially labeled. The proposed methodology generates human-centered explanations by extracting IF-THEN fuzzy rules from the evolving prototypes produced by DISSFCM over successive time intervals. These explanations, expressed in linguistic terms, remain accessible to non-expert stakeholders and are particularly suitable for educational contexts. The Open University Learning Analytics Dataset (OULAD) is utilized for experimentation and validation, providing a realistic scenario for semi-supervised data collection. Visual summaries of the evolving fuzzy rules support the identification of temporal patterns in streaming data. Results show that the model effectively adapts to concept drift while maintaining interpretability. Most notably, it proves robust in handling partially labeled data and variable time granularities, two challenges frequently encountered in real-world Learning Analytics scenarios. The ability to both predict student outcomes and provide intelligible explanations under such constraints highlights the practical value of the approach. To evaluate the quality and relevance of the generated explanations, an expert-based evaluation was conducted. Domain experts evaluated the clarity, usefulness, and accuracy of the explanations in terms of their support for human understanding and decision-making. The results suggest that the explanations were perceived as generally informative and useful, supporting the method’s relevance for human-centered educational applications.

The impact of recent developments in artificial intelligence (AI) on higher education (HE), like other fields, is causing profound transformations that affect the entire organization of academia. A recent systematic review covering scientific papers published between 2016 and 2022, hence only partially covering the ChatGPT phenomenon, highlighted how the use of AI in higher education has grown significantly over the period 2018–2022, with publications increasing two to three times in 2021–2022 compared to previous years. Moreover, research on AI in education have explored how these technologies effectively affect the higher educational context, proving a beneficial impact both in teachers’ and students’ everyday practice. This impact extends beyond the STEM fields, including also courses in humanities. This shift highlights the need for a comprehensive reevaluation of how we approach education in the age of AI, considering both its strengths and limitations across various academic fields. In this paper, we present the results of exploratory research on how the introduction of generative AI (GenAI) in Italian higher education contexts has affected university teaching; results are based on two case studies based on courses in two completely different fields. The first case study concerns the introduction of GenAI in the first-year course of ‘Educational Technologies’ in the program of a Master's Degree in Primary Teacher Education at the LUMSA University; the second case study refers to a third-year ‘Open Data Management’ course in the Computer Science Bachelor’s program at the University of Palermo.

The integration of artificial intelligence (AI) technologies, particularly Large Language Models (LLMs), within educational settings has sparked a notable transformation in the methods of knowledge dissemination and acquisition. This study focuses on using ChatGPT to enhance learning experiences for students with the goal of promoting active learning. The research aims to address the issue of superficial use of AI tools by proposing and implementing innovative methods based on concept maps to effectively leverage these tools, with the aim of fostering active learning and constructive educational experiences for students. The results indicate that students who participated in the experiment gained a better understanding of concepts and performed better compared to their peers who did not take part. Additionally, the study provides insights into students' opinions on various topics, including their perceptions of AI in education, their willingness to use other large language models, their views on the usefulness of ChatGPT in supporting learning, their perceptions of the differences between interacting with ChatGPT and teachers, and their habits in using technological tools.

Young people worldwide use social media. Besides the benefits, such as communication, entertainment, or social support, users also have to deal with negative incidents, such as cyberbullying and its serious consequences. One key factor in mitigating cyberbullying is empathy. Therefore, we developed an empathy training for adolescents, which is led by a virtual learning companion in a social media-like environment and includes direct conversations between the user and the virtual learning companion as well as a video explicating the concept of empathy. This empathy training shall contribute to decreasing users’ bullying intentions and increasing their empathy. Since previous research shows that social media use and cyberbullying are linked to country-specific factors, we evaluated the empathy training in a cross-national experimental study with N = 332 participants from Brazil, Germany, Italy, and Spain. Data were collected via surveys during workshops in schools. Among others, participants answered questions regarding their cyberbullying intentions, levels of cognitive and affective empathy, intentions to help a victim of cyberbullying, and perceived support of the measure. Results show neither decreased bullying intentions nor increased empathy after empathy training. However, cross-national differences were found. Participants’ intentions to bully were significantly higher in Spain and Brazil than in Germany. Furthermore, in the current study, cognitive and affective empathy was significantly higher in the Italian sample than in the German sample. Possibilities for improving the empathy training and the role of cross-national differences are discussed.

Over the past few years, digitalisation has led to the development of new forms of Holocaust memory, with advances in digital technology reshaping and introducing alternative ways of remembering, understanding and representing the Holocaust. The purpose of this study is to examine how three Holocaust survivors - Lily Ebert (100), Gidon Lev (88) and Tova Friedman (85) - share their firsthand experiences on TikTok by segmenting traumatic memories using the platforms' audio-visual aesthetic and adapting their testimonies for the attention spans of young users. Based on 1-year content production and detailed analysis of 84 videos across the three profiles, a mixed-methods approach was applied to identify how each survivor interacts with their 'fans' using a unique communication style and with distinct goals. The results of the multimodal analysis show that the three survivors are engaged in meaningful acts of playful online activism on the memory of the Holocaust by bringing testimony and daily life together, in order to protect historical facts and combat antisemitism and Holocaust distortion.

The HELMeTO 2024 Conference Proceedings highlight the transformative role of Artificial Intelligence (AI) and emerging technologies in reshaping higher education. The 46 accepted research papers explore a wide spectrum of themes including AI in education, digital transformation, inclusion and student well-being, educational policy, ethics, and teacher development. The contributions examine how AI tools foster student engagement, personalize learning, support accessibility, and streamline educational processes, while also addressing critical concerns such as ethical governance, algorithmic bias, and data privacy. Emerging technologies like augmented and virtual reality, game-based learning, and robotics are shown to enhance interactivity and motivation in diverse learning environments. The proceedings also emphasize the importance of inclusive education and mental well-being, calling for holistic approaches that integrate emotional and cognitive development. In parallel, the role of educators is redefined through AI-driven practices, necessitating targeted professional development and ethical training. Policy frameworks such as UNESCO's guidance on generative AI and the European Union's AI Act are examined for their potential to guide responsible implementation.

Predicting whether a newly submitted bug will be resolved quickly or slowly is a crucial aspect of the bug triage process, as it enables project managers to estimate software maintenance efforts and manage development workflows more effectively. This paper proposes a deep learning approach for classifying bug reports into two categories-FAST or SLOW-based on their expected fixing time. The method leverages a feature set composed of the bug description and reporter comments and adopts a transfer learning strategy using pre-trained Large Language Models (LLMs). The problem is framed as a supervised text classification task, where LLMs exploit their ability to learn rich contextual representations of language. We introduce a novel classification workflow that guides the LLM through a structured prompt, combining two design patterns: the persona pattern to contextualize the task and the input semantic pattern to organize textual information. The workflow relies on zero-shot learning to assess whether the intrinsic knowledge embedded in the LLMs is sufficient for this prediction task. We conducted a comprehensive evaluation of three state-of-the-art LLMs across multiple realworld datasets sourced from Bugzilla, encompassing a diverse range of software projects. The experimental results demonstrate that the proposed method is effective in accurately identifying fast-resolving bugs. Among the evaluated models, LLaMA3-8B consistently delivered superior performance. Additionally, the absence of statistically significant performance variations across datasets highlights the generalizability of the approach. Notably, the LLMs maintained strong performance even on small and imbalanced datasets, underscoring their robustness and practical applicability in real-world, data-scarce scenarios.

The widespread use of social media has highlighted potential negative impacts on society and individuals, largely driven by recommendation algorithms that shape user behavior and social dynamics. Understanding these algorithms’ impact is essential but challenging due to the complex, distributed nature of social media networks as well as limited access to real-world data and in particular recommendations, usually not reported. This study proposes to use academic social networks as a proxy for investigating recommendation systems in social media. By employing Graph Neural Networks (GNNs), we develop a model that separates the prediction of academic infosphere (in which a recommender can play a main role) from user behavior prediction, allowing us to simulate recommender-generated infospheres and assess different recommenders’ impact on the model’s performance in predicting future co-authorships. Our approach aims to improve our understanding of recommendation systems’ roles and social networks modeling. To support the reproducibility of our work we publicly make available our implementations: https://github.com/DimNeuroLab/academic_network_project

Code assessment in computer science education is a time-consuming process requiring evaluation of both functionality and coding style. To overcome this problems, we present a novel approach that utilizes Large Language Models (LLMs) for rubric-based code evaluation, addressing the limitations of current automated tools that rely primarily on test cases and code similarity metrics, which sometimes fail to capture more nuanced aspects of student submissions. Our method enables LLMs to analyze student code against instructor-defined rubrics, generating inline, contextualized feedback without altering the original code. This may allow instructors to efficiently review, refine, and finalize grades while maintaining pedagogical oversight. We evaluated the grading performance of five state-of-the-art LLMs (Claude-3.5 Sonnet, GPT-4o, Grok-2, LLaMA-3.3 70B, and DeepSeek-V3) against human graders using a dataset of over 500 coding exercises from an introductory university programming course. Our findings show that Claude-3.5 Sonnet exhibits the highest alignment with human grading, consistently outperforming other models across different exercise types. The dataset and code for reproducibility are available at: https://doi.org/10.5281/zenodo.14879148.

Inclusive education is one of the 17 Sustainable Development Goals that aims to guarantee access and participation to all students considering their learning needs and competencies. Such a goal is particularly relevant in developing countries where socioeconomic conditions make it difficult to provide education, leading to a rate of illiterate adults higher than 99

The complexity behind the analysis of mobile learning activities has requested the development of specifically designed frameworks. When students are involved in mobile learning experiences, they interact with the context in which the activities occur, the content they have access to, with peers and their teachers. The wider adoption of generative artificial intelligence introduces new interactions that researchers have to look at when learning analytics techniques are applied to monitor learning patterns. The task interaction framework proposed in this paper explores how AI-based tools affect student-content and student-context interactions during mobile learning activities, thus focusing on the interplay of Learning Analytics and Artificial Intelligence advances in the educational domain. A use case scenario that explores the framework's application in a real educational context is also presented. Finally, we describe the architectural design of an environment that leverages the task interaction framework to analyze enhanced mobile learning experiences in which structured content extracted from a Knowledge Graph is elaborated by a large language model to provide students with personalized content.

Appears in: EDULEARN23 Proceedings Publication year: 2023Pages: 6160-6167ISBN: 978-84-09-52151-7ISSN: 2340-1117doi: 10.21125/edulearn.2023.1602Conference name: 15th International Conference on Education and New Learning TechnologiesDates: 3-5 July, 2023Location: Palma, Spain