It is challenging for autistic people to interpret and respond appropriately to social scenarios due to differences in emotion understanding and recognition abilities. This paper aims to design, implement, and evaluate an adaptive, gamified tool to support emotion recognition skill development in children with ASD. First, we propose a novel adaptive algorithm that facilitates a personalized progression from avatar to human faces according to the user’s real-time performance accuracy. Second, we present the implementation of EmoBridge, a mobile application that integrates this algorithm within a gamified learning session and includes a browse section for users to view emotion-categorized images and upload their own. Third, we report findings from a preliminary experiment involving 18 children aged 7-12, including both neurotypical and ASD-diagnosed participants, and their parents. The results demonstrate EmoBridge’s potential to support emotion recognition, particularly for verbal chi ldren with ASD, while also revealing specific limitations for non-verbal users. The feedback collected provides valuable insights that will guide future iterations and research, contributing to the development of more inclusive and adaptive assistive technologies.

As online learning becomes increasingly prevalent, it is essential to understand students' perspectives and address the challenges they encounter. The developing system, ScaffoldiaMyMaths, aims to support the mathematics learning of underprivileged and lower-ability primary school students. This system integrates context-specific and dynamic scaffolding to provide personalized support, particularly in the study of fractions. In addition to standard learning management features, ScaffoldiaMyMaths offers dynamic virtual manipulative representations and adaptive scaffolding for problem-solving, complemented by Al-based real-time feedback. The Al-generated feedback enhances the learning experience by delivering immediate, tailored responses to students' queries. It analyzes students' inputs to offer precise guidance, suggest alternative problem-solving strategies, and identify areas needing further practice. By utilizing Al, ScaffoldiaMyMaths ensures that feedback is both prompt and customized, effectively addressing individual learning gaps and fostering a deeper understanding of mathematical concepts.

Beginner programmers can develop an intuitive understanding of programming by leveraging the motivating field of multimedia to visually inspect outputs and experiment with different ways to solve problems. This paper presents MediaLib, a Python library designed to facilitate multimedia programming and lessen the cognitive load associated with programming for novice programmers. In addition, we designed an official MediaLib website which contains the library itself, two tutorials, and clear documentation. The tutorial clearly presents the learning objectives of each lesson and contains exercises related to MediaLib. We designed these exercises to help students gain knowledge incrementally, without requiring in-depth maths knowledge.

Integrating transformers with graph representation learning has emerged as a research focal point. However, recent studies showed that positional encoding in Transformers does not capture enough structural information between nodes. Additionally, existing graph neural network (GNN) models face the oversquashing issue, impeding information retention from distant nodes. To address, we transform graphs into regular structures, such as tokens, to enhance positional understanding and leverage transformer strengths. Inspired by the visual transformer (ViT) model, we propose partitioning graphs into patches and apply GNN models obtain fixed size vectors. Notably, our approach adopts contrastive learning for in-depth graph structure and incorporate more topological information via Ricci curvature to alleviate over-squashing problem by attenuating the effects of negatively curved edges while preserving the original graph structure. Unlike existing graph rewiring methods that directly modify graph structure by adding or removing edges, this approach is potentially more suitable for applications such as molecular learning where structural preservation is important. Our innovative pipeline subsequently introduces the PerformerMixer, a transformer variant with linear complexity, ensuring efficient computation. Evaluations on real-world benchmarks demonstrate our framework’s superior performance, like Peptides-func and achieve 3-WL expressiveness.

The STEAM approach in education has been gaining increasing popularity over the last decade. This is due to its potential in enhancing students' learning, when teaching arts and scientific disciplines together. This paper introduces the PICA-PICA concept, where we aim to develop a smart customisable environment, combining, in a unique way, teaching programming in conjunction with the engineering of artworks. The PICA-PICA concept was implemented, used and tested in real-life, by upper primary school students in Japan, during a 4-day workshop. Initial results illustrated the quality of the solution proposed by PICA-PICA. We noted that the integration was perceived as smooth, and not contrived: all participants understood how to use the PICA-PICA environment to engineer programmable art objects. Furthermore, the PICA-PICA approach led to high motivation: children did not get bored and were fully engaged. Finally, the quality of their work as a learning outcome was high: by including a programming segment with the other expressive activities in the artwork, the children were able to design the electronics in a more concentrated and meaningful way than their curriculum-structured learning. This study also presents an innovative implementation of the STEAM approach using Micro:bits technology to create exciting artwork whilst using household recyclable items, which also teaches about sustainability. The involvement of parents and their interest in learning is another unique aspect of this study.

This research proposes to develop an adaptive ontology-based learning support system for computer programming learning. Firstly, the system adopts a previously developed ontology called CONTINUOUS, which represents programming concepts and their relation in a graph and makes use of its content to serve as hints within programming questions. Secondly, we design an adaptive strategy for recommending suitable exercises to learners, which uses CONTINUOUS as metadata of exercises and the Elo rating system to estimate learners' skills. This work aims to design a system to provide personalized exercise path for the learner and evaluate its effectiveness.

In this paper, we present an ontology-based visualization support system for e-book learners, which provides not only a meaningful receptive learning environment but also a meaningful discovery learning environment. Those two environments are developed to help e-book learners to effectively construct their knowledge frameworks. A series of experiments were conducted on four undergraduate classes instructed by two professors (A and B): two classes(one guided by A and the other guided by B) were assigned as control groups and studied with one e-book chapter in receptive learning environment while another two classes (one guided by A and the other guided by B) were assigned as experimental groups and studied with the same e-book chapter in discovery learning environment. For analyzing the learner behavior, K-means clustering algorithm is performed not only by considering the number of total command actions and the cumulative duration of stay on target pages as learner features, but also by considering the duration of stay on each target page (in total 15 pages) as learner features. Learners’ behavior differences in e-book system are examined and discussed.

For the purpose of exploring strategies for multimedia learning object (LO) suggestion in customizable language learning support systems, particularly suggestion for learners with visual and verbal cognitive style preferences, a learning style based experiment was conducted in a Japanese grammar course. The support system under examination offers two learner modes: Open mode, which provides learners with both visual and verbal LOs, and Style-Matching mode, which provides visual learners with only visual LOs and verbal learners with only verbal LOs. Ninety students were assigned to three groups on the basis of their visual/verbal learning styles preferences and their previously measured learning achievement. Experimental group A studied with Open mode; experimental group B studied with Style-Matching mode; and the control group studied with verbal LOs from the course textbook. Learning performance differences among the three groups were examined in terms of (a) learning perception (including technology acceptance measures, cognitive load and satisfaction with learning mode) and (b) achievement differences. The control group had significantly lower scores on visual tasks than experimental groups, but no significant differences in scores were found on verbal tasks among these three groups. Moreover, despite the learners’ report of significantly higher distraction in Open mode, the learning motivation of learners with stronger visual style preference improved more in Open mode than in Style-Matching mode.

In this paper, we present a system framework making use of e-book logs for visualization learning support systems intended to provide meaningful learning environment for e-book learners. An ontology-based visualization support system, which supports not only meaningful reception learning but also meaningful discovery learning, is designed and developed to help e-book learners to effectively construct their knowledge frameworks. Three main functions are provided in this personalized visualization support system: (1) for any knowledge of E-books, the learner can get its e-book location information and a relation map including its relevance knowledge and their relations; (2) for any page range of any e-book, the learner can get a relation map including knowledge shown in those pages, their relevant KPs and their upper concepts; (3) for any learning period, the learner can check the knowledge map including the knowledge they had read, the relevance knowledge and their relations. Compared to the passive reception environment, in meaningful discovery learning environment learners are encouraged to actively locate new knowledge in their own knowledge framework and restructure existing knowledge by detecting hidden relations between relevant KPs through reflecting the attributes of acquired knowledge visually. Meanwhile the iterative procedure of confirmation and modification in their own relation map ensures that they check the logical consistency of their ideas and clear up misunderstandings.

In this paper, we investigate the variation of learning attitude and motivation under the environment of a language learning support system. The learner data involve 60 participants' attitude/motivation rankings before and after studying target grammar contents with the support of the system. Significantly improvement in attitude and motivation was suggested by the analysis. Furthermore, considering learning attitude and motivation as not only conditions but also outcomes for learning, we analyze the relations among learning attitude, motivation, perceptions and achievement.

In this paper, we investigate, the learning attitude, motivation and learning perceptions (including the Satisfaction for learning mode, technology acceptance, mental effort and mental load) of two groups of learners who studied with two different modes of a learning support systems. Group A of 31 students studied with Open Mode which provides learners with both visual and verbal learning objects while group B of 34 students studied with Style-Matching Mode which provides visual learners with only visual learning objects and verbal learners with only verbal learning objects. The learning attitude, motivation, and learning perceptions of these two groups are investigated and analyzed from learning style perspective.

In this paper, we investigate, from learning style perspective, the main factors that affect the learning performance of the learner while using learning support systems. An experiment was conducted to evaluate two different modes of a customizable language learning support system. Students in experimental group A, who were provided with both visual and verbal learning objects, had more difficulty to focus on study compared to those in experimental group B, who were provided with only the learning objects matching their learning style while both using the system. Moreover, 53.3% of students in experimental group A believed the type of LOs, which they preferred more and felt more comfortable with, was not the type of LOs which was more effective for their learning.

This paper describes a course-centered ontology of a specific Japanese grammar course. The leaf concepts of this course-centered ontology represent the grammar points (in total 205 grammar points), and the relations between them include concept dependences, similarities and contrasts, and even grammatical equivalence phenomena. A customizable learning support system based on this course-centered ontology has also been built to help instructors organize the course teaching materials and provide the learners personalized learning objects.