Learning Domain Research Articles

Design Insights for Developing Deepfakes Awareness Games for Adolescents

Around a decade ago, significant advancements in the deep learning domain like Generative Adversarial Networks (GANs) started to facilitate the creation of new synthetical content. The development of deepfake algorithms gained momentum due to the increase in computing power and open-source available resources which facilitated broader access to advanced deep learning methods. While initially deepfakes were created by creative pursuits in various media and marketing campaigns, they started to be misused driven by malicious purposes in various social media manipulation campaigns. These campaigns imply spreading disinformation and misinformation for discrediting individuals and creating confusion in relation to specific topics and events and are meant to altering users’ beliefs and behaviours. Such facts raised important ethical and societal concerns. Although deepfakes still represent a recent research area, a large body of studies is dedicated to, on the one hand, generation, and detection of deepfakes, and on the other hand, to understanding their implications and consequences for society. At the same time, governmental and practitioner efforts are devoted to containing their use and impact on society by proposing various strategies and programs. Nevertheless, both academic and societal efforts are in an incipient stage and often adopt a generalist perspective. This represents a crucial point in relation to the fact that many users in various social media platforms like TikTok is represented by adolescents that are known to be unaware and more vulnerable to media content in general, and deepfakes content, in particular. This represents the knowledge gap that this research aims to tackle by synthesizing design insights as requirements and guidelines that can be used when developing and deploying deepfakes awareness games for producing and/or enhancing awareness of adolescents. Accordingly, a systematic literature review is conducted and merged with previous experience in this domain by adopting a transdisciplinary research approach that merges methods and techniques from the deepfakes, AI, social media, cyber security, and gamification domains.

Digital Drone Education in Sarawak: Enhancing STEM Learning through Hands-on Training for School Students

In Sarawak, concerns about low STEM enrolment, digital inclusivity gaps for rural children, and the transition from digital consumption to digital creation led to the introduction of the Drone Education Program. Participants began with foundational drone operation training, advancing to capturing images and videos. Subsequently, they were acquainted with elementary photo and video editing techniques, enabling them to enhance their drone-generated content. The aim of this research is to gauge the impact of the Drone Education Program on refining drone operations, bolstering digital editing competencies, and stimulating STEM engagement in students. In our methodology, we employed tailored pre- and post-surveys for evaluation, and younger students provided feedback via an emoji-based survey. Further analytical methods encompassed non-parametric tests, Exploratory Factor Analysis (EFA), and linear mixed-effect modelling. Results indicated significant post-intervention improvements in the targeted skills. The EFA grouped 21 Likert items into five distinct learning domains. Notable insights include district-wise variances in learning outcomes and the influential role of gender. The initiative showcases promise in confronting Sarawak's educational barriers, championing digital inclusivity, and preparing a digitally proficient generation.

A systematic review of flipped classroom approaches in language learning

This systematic literature review aims to offer an in-depth analysis of the existing research landscape regarding the application of the flipped classroom approach within the domain of language learning. The investigators methodically selected and analyzed 57 papers from 656 documents using a systematic and rigorous approach to literature selection. Our findings demonstrate the significant international impact of this field, with contributions from many countries and districts as well as globally excellent academic institutions. This review examines multiple areas of research, including improving students’ perceptions and attitudes in flipped classrooms, language proficiency, motivational factors, etc. In addition, the research methods employed were evaluated, with a particular emphasis on mixed methods which combined the quantitative method with the qualitative method. Even though the application of flipped classrooms in language learning has achieved some achievements, some limitations are obvious. The most notable are the problem of sample size, the difficulty of technology integration, the lack of teacher training, the singularity of language research, etc. The research gaps revealed in this literature review, indicate interesting possibilities for future research. The diversity of research languages, the multiplicity of research methods, the integration of advanced technologies, and intensive teacher training will be significant of the area focus. This systematic literature review not only enriches the existing knowledge system but also has the potential to the direction of the development of flipped classrooms in language learning.

The Multiple Millionaires' Problem: New Algorithmic Approaches and Protocols

We study a fundamental problem in Multi-Party Computation, which we call the Multiple Millionaires Problem (MMP). Given a set of private integer inputs, the problem is to identify the subset of inputs that equal the maximum (or minimum) of that set, without revealing any further information on the inputs beyond what is implied by the desired output. Such a problem is a natural extension of the Millionaires Problem, which is the very first Multi-Party Computation problem that was presented in Andrew Yaos seminal work (FOCS 1982). A closely related problem is MaxP, in which the value of the maximum is sought. We study these fundamental problems and describe several algorithmic approaches and protocols for their solution. In addition, we compare the performance of the protocols under several selected settings. As applications of privacy-preserving computation are more and more commonly implemented in industrial systems, MMP and MaxP become important building blocks in privacy-preserving statistics, machine learning, auctions and other domains. One of the prominent advantages of the protocols that we present here is their simplicity. As they solve fundamental problems that are essential building blocks in various application scenarios, we believe that the presented solutions to those problems, and the comparison between them, will serve well future researchers and practitioners of secure distributed computing.

The perception of complex engineering problems among engineering teachers and students: A case study

The complex engineering problems (CEPs) are characterized by challenges such as: a) involving extensive or inconsistent, incompatible data from engineering and other offshoots, b) unable to be solved in the absence of in-depth engineering knowledge. The engineering teachers are not found successful in the design of a CEP because it does not touch the real issues from industry and / or environment. Therefore, the expertise of applicable skill cannot be evaluated and the reason behind this failure is the lack of or poor intellectual perception of the attributes of CEPs among the teachers. In order to help teachers, the relation between attributes, knowledge profiles and CEPs is elucidated in this work by giving examples. The main objective of this article is to establish an understanding of the student’s level of perceiving a complex engineering problem with respect to the learning domain. The Survey was conducted among 70 participants and the quantitative approach employed, utilizing a questionnaire survey and interviewing teachers to gather data from respondents selected through random probability sampling. It is concluded that Industrial training (like apprenticeships), laboratory practices and final year design project are good tools for the complex engineering activities/ exercises. The results obtained in the case study were found promising, therefore, it is recommended to extend this study by using various engineeringinstitutions to ascertain a framework on CEP solving. skills.

Interpretable Transformer Neural Network Prediction of Diverse Environmental Time Series Using Weather Forecasts

AbstractTransformer neural networks (TNNs) have caused a paradigm shift in deep learning domains like natural language processing, gathering immense interest due to their versatility in other fields such as time series forecasting (TSF). Most current TSF applications of TNNs use only historic observations to predict future events, ignoring information available in weather forecasts to inform better predictions, and with little attention given to the interpretability of the model's use of explanatory inputs. This work explores the potential for TNNs to perform TSF across multiple environmental variables (streamflow, stage, water temperature, and salinity) in two ecologically important regions: the Peace River watershed (Florida) and the northern Gulf of Mexico (Louisiana). The TNN was tested and its prediction uncertainty quantified for each response variable from one‐to fourteen‐day‐ahead forecasts using past observations and spatially distributed weather forecasts. A sensitivity analysis (SA) was performed on the trained TNNs' attention weights to identify the relative influence of each input variable on each response variable across prediction windows. Overall model performance ranged from good to very good (0.78 < NSE < 0.99 for all variables and forecast horizons). Through the SA, we found that the TNN was able to learn the physical patterns behind the data, adapt the use of input variables to each forecast, and increasingly use weather forecast information as prediction windows increased. The TNN's excellent performance and flexibility, along with the intuitive interpretability highlighting the logic behind the models' forecasting decision‐making process, provide evidence for the applicability of this architecture to other TSF variables and locations.

EVALUATING THE DENTAL LEARNING ENVIRONMENT: A STUDENT-CENTRIC PRESPECTIVE ACROSS PRECLINICAL AND CLINICAL YEARS USING DUNDEE READY EDUCATION ENVIRONMENT MEASURE (DREEM) TOOL

Objectives: The learning environment has a profound role in students learning and success. It also has an integral role in the development of the curriculum. The aim of this study to evaluate the learning environment and compare dental students' perception of the learning environment of pre-clinical and clinical years. Materials and Methods: This cross-sectional study was conducted from October 2023 to April 2024. A total of 170 both male and female BDS students from all four years at the Bacha Khan College of Denistry were included in the study. A stratified random Sampling Method was used. Students migrated from other colleges in the middle of the sessions and students currently working as interns or house officers were excluded from the study. Dundee Ready Education Environment Measure (DREEM), a 50-item questionnaire that uses a 5-point Likert scale (4 = strongly agree to 0 = strongly disagree) was used to measure strength of agreement with statements on five domains of learning. Learning (L), Teaching (T), Academic self-perception (ASP), Atmosphere (A) and Social self-perception (SSP). Frequency distributions and descriptive statistics were tabulated in IBM SPSS (version 21). Mean scores for each dimension of learning were used for analysis. Comparative analysis was done using independent sample t test. Results: The DREEM questionnaire was completed by 147 students, 78 1st and 2nd year students (preclinical training phase) while 69 3rd and 4th year students. 34.6% of the participants were male and 66.6% were female. The mean age of the participants was 21.4 years (SD=1.72). Tukey's post hoc analysis within the participants revealed a significantly lower DREEM score of the Final year compared with the 1st year (p < .003), and 3rd year (p< 0.031). The items within each domain with the biggest signifi cant differences (p ≤ 0.05) was between Pre-Clinical Training Phase and Clinical training. Conclusion: This study examined the educational setting in our institution. The students of Bacha Khan College of Dentistry showed a predominantly satisfaction with their educational environment. The primary results indicated that students in the early stages of their studies, namely those in their pre-clinical years had positive opinions across all aspects of their learning environment. Overall, the findings from the study highlight the importance of understanding how students perceive their educational environment throughout diff erent stages of their dental training. By delving deeper into these differences and correlations, educators and administrators can make informed decisions to enhance the curriculum and ultimately improve the learning experience for students. This implies that a more tailored and effective approach to dental education could lead to better outcomes and preparedness for future dental professionals.

Evaluating deep neural networks in optimizing drug discovery and precision medicine: A review

Introduction/Background: Deep neural networks have shown great promise in advancing drug discovery and precision medicine. By leveraging large amounts of complex biomedical and chemical data, deep learning approaches can identify novel targets, predict drug-target and drug-drug interactions, generate new molecular structures, and assist in personalized treatment selection and development. However, fully utilizing deep learning techniques for optimization across the drug development pipeline remains an ongoing challenge. Materials and Methods: A comprehensive literature review was conducted using major bibliographic databases including PubMed, Web of Science, and Scopus. Search terms included combinations of "deep learning", "drug discovery", "precision medicine", "biomedical data", and "neural networks". Over 200 papers published between 2010-2023 related to deep learning applications in pharmacology and genomics were identified and reviewed. Results: Deep learning has been widely applied at various stages of the drug discovery process including target identification/prioritization, lead generation/optimization, and prediction of molecular properties. Convolutional neural networks are commonly used for the representation and classification of biological sequence and image data for tasks such as gene expression analysis and pathogen detection from microscopy images. Graph neural networks effectively model compound structures and interactome networks to predict molecular bindings and disease associations. Multi-modal neural networks integrate diverse data types for personalized treatment response prediction and biomarker discovery. Challenges remain around data and model interpretation, generalization to new targets/diseases, and integration across domains. Discussion: While deep learning has shown promise, rigorous benchmarking and validation on real-world clinical endpoints are still needed to establish usefulness in decision-making. Data and model transparency must be improved to enable scientific insights. Privacy and security risks accompanying "real world" biomedical big data will require ethical practices. Standardization and sharing of resources/protocols could accelerate progress by enabling comparison of techniques. Combining deep learning with other AI paradigms like causal inference may further improve utility in drug discovery and precision healthcare. Conclusion: Deep neural networks demonstrate potential for optimizing drug development and precision medicine applications. Continued advancement relies on addressing challenges around data, models, validation, and ethics. Multi-disciplinary collaborations integrating machine learning, molecular biology, medicine, and other domains are needed to fully realize benefits to patients.

IMPLEMENTATION OF ONLINE ASSESSMENT FOR PSYCHOMOTOR DOMAIN IN ENGINEERING PROGRAM: AN INSIGHT FROM STUDENT’S PERSPECTIVE

Assessment plays an essential role in learning and must be designed to measure the student's learning and align with the intended learning outcomes. Prior to the pandemic, most of the assessments were performed face-to-face. However, during the pandemic, the assessment could not be conducted as face-to-face activities, which caused challenges to the assessment that required face-to-face implementation. This study seeks to critically examine the implementation of the online assessment conducted during the COVID-19 pandemic to assess the psychomotor learning domain. The study used a qualitative case study that utilized document analysis and interview protocol. Purposive sampling was adopted, and the sample population consists of 60 students taking the course during the March – July 2020 semester. The data collection procedure was separated into Stages 1 and 2, Document Analysis and Semi-structured in-depth interviews analyzed using thematic analysis. The findings show that certain conducted assessments are unsuitable for assessing the psychomotor learning domain due to the nature of the assessment and the criteria adopted to assess the students. It is suggested that the implementation of assessment in assessing students' psychomotor skills through online be re-examined. This study also found that the students had to face four main challenges in conducting the online assessment that could affect the students’ performance: internet connectivity, accessibility to a decent device to complete the assessment, lack of explanation on the assessment and the environment while conducting the assessment.

The Changing Educational Landscape for Sustainable Online Experiences: Implications of ChatGPT in Arab Students’ Learning Experience

The integration of artificial intelligence (AI) into education is rapidly transforming traditional paradigms of teaching and learning. ChatGPT, an advanced AI system capable of sophisticated language generation, represents a potentially disruptive innovation in this sphere. This study employed a descriptive cross-sectional design using a cross-sectional approach to examine ChatGPT’s emerging role in reshaping students’ learning experiences within higher education. A survey of 473 undergraduate students at King Faisal University, Saudi Arabia assessed ChatGPT’s usage patterns, effectiveness, and implications across diverse learning domains. Data were collected through a structured online survey distributed over a month and analyzed using quantitative measures (i.e., SPSS). Sixty precent (60%) of the students reported using ChatGPT for homework assistance, with 27.7% doing so weekly. Additionally, 45.5% utilized it for research, 35.5% for exam preparation, and approximately 30% for language learning and content discussions. Over 60% of the students recognized moderate to significant positive impacts of ChatGPT on their academic performance, research skills, and technical competencies. However, 38.7% reported no discernible effects on critical thinking. Factor analysis identified engagement as a key pathway for ChatGPT to enrich learning, with usage strongly linked to participation and motivation. Regression analysis determined usage frequency, satisfaction, and perceived usefulness as significant predictors of ChatGPT’s academic effectiveness, evidenced by beta coefficients of 0.21 for usage frequency and 0.25 for satisfaction. The findings showcase ChatGPT’s promising versatility as an AI-powered educational tool, while emphasizing the need for ethical frameworks, balanced expectations, and holistic adoption strategies to maximize benefits and minimize risks. Further research into pedagogical integration capabilities and evolving impacts will be critical as advanced AI becomes increasingly pervasive throughout education.

Effective Teaching for Classroom Management Workshop: The Ghanaian Teachers’ Perspectives and Evaluation

The paper is about teachers’ reflection on and evaluation of continuous professional training and development. It is a qualitative survey, grounded in narrative inquiry framework of educational research with the assumptions that humans are social storytelling organisms. With a convenient sampling technique, 619 responses were received from 778 participants. Adopting a participant-centred mode of teaching, participants evaluated the workshop in their own words. With a human cognition of thinking, learning, and understanding, comments were classified into Bloom’s (1956) three learning domains of cognitive, affective, and skills combined with Perry & Booth’s (2021) concept of evaluating facilitators and workshop organizations and Kirkpatrick’s (1998) four levels of evaluation. Participants insisted that this workshop be extended to other schools to build the required foundation for students. They mentioned also that their exposures to contemporary instructional strategies meet the globally acceptable standards and are challenged as 21st century teachers. Participants maintained that they would go back to the classroom with different mind-sets to include learners during lesson planning.

Deep shared learning and attentive domain mapping for cross-domain recommendation

Cross-domain recommendations (CDR) present a viable solution and are increasingly used to address the cold-start problem. Recently, CDR methods are utilizing deep models to generate latent preferences from context vectors or rating matrices and transfer these preferences between domains. However, many of these models focus on learning latent preferences using domain-related information and often disregard preference patterns from the contrary domain. Incorporating the contrary domain preference patterns into deep models can improve the generation of more effective latent representations. Moreover, existing CDR models face challenges in effectively transferring mapped preferences between domains due to the large features disparity between them. In this study, we tackle these problems and present a novel Deep Shared Learning and Attentive Domain Mapping (DSAM) approach for CDR. Specifically, we propose a variant of Long Short-Term Memory (LSTM) called shared learning LSTM, which incorporates the learning of cross-domain preference patterns alongside domain-specific user/item embeddings derived from textual reviews to dynamically generate shared contextual representations in each domain. We further exploit a multi-head self-attentive network to match item-specific knowledge from the source and target domains into different subspaces. We aggregate this learned knowledge to predict rating scores for cold-start users in the target domain. We efficiently optimize this framework in an end-to-end fashion. Experimental results on five real-world datasets demonstrate the effectiveness of our proposed approach against various groups of recommendation models. Additionally, we provide insights to help understand the model architecture and its robustness in handling cold-start users.

Deep learning domain adaptation to understand physico-chemical processes from fluorescence spectroscopy small datasets and application to the oxidation of olive oil

Fluorescence spectroscopy is a fundamental tool in life sciences and chemistry, with applications in environmental monitoring, food quality control, and biomedical diagnostics. However, analysis of spectroscopic data with deep learning, in particular of fluorescence excitation-emission matrices (EEMs), presents significant challenges due to the typically small and sparse datasets available. Furthermore, the analysis of EEMs is difficult due to their high dimensionality and overlapping spectral features. This study proposes a new approach that exploits domain adaptation with pretrained vision models, along with a novel interpretability algorithm to address these challenges. Thanks to specialised feature engineering of the neural networks described in this work, we are now able to provide deeper insights into the physico-chemical processes underlying the data. The proposed approach is demonstrated through the analysis of the oxidation process in extra virgin olive oil (EVOO), showing its effectiveness in predicting quality indicators and identifying the spectral bands and thus the molecules involved in the process. This work describes a significantly innovative approach to deep learning for spectroscopy, transforming it from a black box into a tool for understanding complex biological and chemical processes.

Effect of Self-Regulated Learning (SRL) on Academic Achievement of Students in the Subject of Chinese Language

The study was primarily focused on investigating the effect of self-regulated learning (SRL) on the academic achievement of Chinese language learners among non-native learners. It was a true experimental research methodology based on pre-test and post-test research design. It was conducted at the University of Central Punjab (UCP) Lahore by selecting 70 students. After conducting the pre-test it was randomly allocated into dual groupings. One is the experimental while the other is the control group. After 16 weeks of treatment, a post-test was conducted. The results showed that the higher the achievement of learners of the experimental group than the learners alienated into the control group of the Chinese language post-test although all students performed equally in the pre-test which exhibited that all students had the same knowledge about the Chinese language before conducting research. It is also recommended that teachers should be conscious of employing the stratagem that initiates self-regulating learning in accordance with the contents of the textbook. Foregoing the above, it is suggested that the developers responsible for implementing the curriculum should also design the courses to inspire students to boost self-regulating-learning skills among them. Refresher-Courses for the professional-cum-novice instructors should be conducted to include the significance of self-regulating learning for students as well. The present research study was focused on the behavioral domain of self-regulating learning with respect to academic achievement of students learning the Chinese language at the university level and more is appreciated to fill the gap.

Exploring ESL Learners’ Reading Strategies Based on the Theory of Reciprocal Determinism

The purpose of this study is to explore the relationship between reading strategies and Bandura’s (1986) theory of Reciprocal Determinism. Previous research acknowledged the concept of Reciprocal Determinism as an effective teaching approach, however, limited research has been done to consider how reading is influenced by learners’ personal factors, behavior and environment. Based on four Research Questions, the study explores the perceived use of reading strategies by 225 undergraduates who completed a critical reading course at a local university. These learners of English as a Second Language responded to an online questionnaire adapted from Mokhtari and Sheorey’s (2002) Survey of Reading Strategies. Based on the learners’ ratings of the reading strategies, namely Global Reading Strategies, Problem-Solving Strategies and Support Strategies, the findings revealed that there is a significant relationship between the learners’ perceived use of reading strategies and the three factors of Reciprocal Determinism, namely Personal Factors, Behaviour and Environment indicating a cyclical effect that infuences learners’ reading comprehension. This exploration of reading strategies through this theory has pedagogical implications on how reading is taught to scaffold and help learners increase metacognitive awareness and self-efficacy. Overall, this study contributes to a further understanding of the importance of taking into account the three factors in a reading environment and the relationship between reading strategies and RD. Future studies could be an extension of how the cyclical effects of Reciprocal Determinism and reading strategies could be strategically integrated into the teaching of skilled and unskilled readers and in other learning domains.

Machine learning and domain decomposition methods - a survey

Hybrid algorithms, which combine black-box machine learning methods with experience from traditional numerical methods and domain expertise from diverse application areas, are progressively gaining importance in scientific machine learning and various industrial domains, especially in computational science and engineering. In the present survey, several promising avenues of research will be examined which focus on the combination of machine learning (ML) and domain decomposition methods (DDMs). The aim of this survey is to provide an overview of existing work within this field and to structure it into domain decomposition for machine learning and machine learning-enhanced domain decomposition, including: domain decomposition for classical machine learning, domain decomposition to accelerate the training of physics-aware neural networks, machine learning to enhance the convergence properties or computational efficiency of DDMs, and machine learning as a discretization method in a DDM for the solution of PDEs. In each of these fields, we summarize existing work and key advances within a common framework and, finally, discuss ongoing challenges and opportunities for future research.

Artificial intelligence and machine learning in corporate governance: A bibliometric analysis

BACKGROUND: The study deeply explores the thriving domains of artificial intelligence (AI) and machine learning (ML) in corporate governance. OBJECTIVE: The study aims to thoroughly examine the rapidly developing fields of artificial intelligence (AI) and machine learning (ML) in corporate governance. METHODS: After completing an in-depth analysis of 229 research studies published between 2008 and 2023 (using software tools such as RStudio, VOSviewer, and Excel),), the study reveals a notable increase in publications since 2022. Corporate social responsibility (CSR), environmental, social, and governance (ESG) issues, executive remuneration, and sustainability are all considered as important key focal areas of focus. Scholars in this field are notably at the forefront from Taiwan, the United States, and China. IMPLICATIONS: However, the study stress the necessity for further researches to estimate the efficacy of different AI and ML methodologies. This may guide evidence-based governance practices various industries and geographical areas.

Leading the way: implementing aged care innovations.

The objective of this study was to identify the key characteristics of leaders that support the implementation of innovations in aged care settings. We conducted a secondary analysis of papers from a large scoping review that identified how leaders supported the implementation of innovations in aged care. Once imported into NVivo12, the findings were deductively coded using the domains of Bloom's taxonomy of learning. Each parent code was then inductively analyzed to identify key characteristics within each domain. Our review identified four types of knowledge, five skills, and six attitudes that leaders should exhibit to better support the implementation of innovations within aged care settings. In addition to our findings regarding Bloom's learning domains, we identified nine leadership behaviors that participants in the included papers perceived as valuable for enhancing the implementation process. Furthermore, we identified four key organizational elements that support leaders in navigating and facilitating the implementation of innovations within aged care settings. Our review identified the characteristics that leaders should demonstrate when supporting the implementation of innovations in aged care. Importantly, our findings also emphasized the changing role of leadership from a hierarchical approach to a more collaborative, supportive, and empowering style. The insights identified in this review will help to guide aged care leaders, stressing the significance of adaptable and relational leadership styles that will guide the implementation of innovations within the aged care sector. http://links.lww.com/IJEBH/A271.

SiGNN: A spike-induced graph neural network for dynamic graph representation learning

In the domain of dynamic graph representation learning (DGRL), capturing the temporal evolution within real-world networks is of paramount importance. Spiking Neural Networks (SNNs), renowned for their temporal dynamics and low-power characteristics, provide an efficient solution for temporal processing in DGRL tasks. However, due to the spike-based information encoding mechanism of SNNs, existing DGRL methods that employ SNNs face significant limitations in their representational capacity. To address this issue, we propose an innovative framework named Spike-induced Graph Neural Network (SiGNN) for learning enhanced spatiotemporal representations on dynamic graphs. Specifically, we achieve a harmonious integration of SNNs and GNNs through a novel Temporal Activation (TA) mechanism. This mechanism enables SiGNN to effectively harness the temporal dynamics of SNNs while circumventing the representational constraints imposed by the binary nature of spikes. Furthermore, leveraging the inherent adaptability of SNNs, SiGNN incorporates an in-depth analysis of evolutionary patterns within graphs across multiple time granularities, thereby facilitating the acquisition of multiscale temporal node representations. Extensive experiments on various real-world dynamic graph datasets demonstrate the superior performance of SiGNN in the node classification task. Our code is publicly available at https://github.com/CharliedoD/SiGNN.

The Transformative Teaching Practice of Teachers and Cultivation of 21st Century Skills of College Students in Selected Sports Universities in China

This study aims to explore the impact of teachers' transformative teaching practices on college students' mastery of 21st-century skills, with a focus on the teaching dimensions of "Personalized Considerations" and "Inspiring People," as well as the students' mastery of "Professional Skills" and "Life Skills." By analyzing students' assessments of their proficiency in 21st-century skills, the research demonstrates a significant positive correlation between transformative teaching behaviors and students' skill mastery. Specifically, the ability of teachers to motivate and provide personalized support significantly enhances students’ performance in learning, life, literacy, and professional domains. The data analysis revealed that students generally rated their mastery of 21st-century skills, including learning, life, literacy, and professional skills, as "agree," indicating a moderate level of manifestation. Additionally, differences in students' skill levels across gender, age, and grade level were found to be insignificant. However, the impact of transformative teaching practices on students' skill development was notably significant, particularly in the areas of "Charm Influence" and "Intellectual Stimulation." Regression analysis further highlighted that "Personalized Considerations" and "Inspiring People" teaching styles are strongly associated with students' mastery of 21st-century skills, especially in enhancing "Life Skills" and "Professional Skills." The proposed intervention program, developed from this study, aims to strengthen teachers' transformative teaching abilities to improve students' mastery of professional and life skills. Through personalized support and motivational guidance, college students will be better equipped to face future challenges in both their personal and professional lives. This study provides practical recommendations for educators to enhance teaching strategies to more effectively cultivate students' core 21st-century skills.