Search Results (33)

This paper presents a scalable and low-cost Retrieval Augmented Generation (RAG) architecture designed to enhance learning in university-level courses, with a particular focus on supporting students from economically disadvantaged backgrounds. Recent advances in large language models (LLMs) have demonstrated considerable potential in educational contexts; however, their adoption is often limited by computational costs and the need for stable broadband access, issues that disproportionately affect low-income learners. To address this challenge, we propose a lightweight, mobile, and friendly RAG system that integrates the LLaMA language model with the Milvus vector database, enabling efficient on device retrieval and context-grounded generation using only modest hardware resources. The system was implemented in a university-level Data Mining course and evaluated over four semesters using a quasi-experimental design with randomized assignment to experimental and control groups. Students in the experimental group had voluntary access to the RAG assistant, while the control group followed the same instructional schedule without exposure to the tool. The results show statistically significant improvements in academic performance for the experimental group, with p < 0.01 in the first semester and p < 0.001 in the subsequent three semesters. Effect sizes, measured using Hedges g to account for small cohort sizes, increased from 0.56 (moderate) to 1.52 (extremely large), demonstrating a clear and growing pedagogical impact over time. Qualitative feedback further indicates increased learner autonomy, confidence, and engagement. These findings highlight the potential of mobile RAG architectures to deliver equitable, high-quality AI support to students regardless of socioeconomic status. The proposed solution offers a practical engineering pathway for institutions seeking inclusive, scalable, and resource-efficient approaches to AI-enhanced education. Full article

Artificial Intelligence (AI) has brought a revolution in many areas, including the education sector. It has the potential to improve learning practices, innovate teaching, and accelerate the path towards personalized learning. This work introduces Reinforcement Learning (RL) methods to develop a personalized examination scheduling system at a university level. We use two widely established RL algorithms, Q-Learning and Proximal Policy Optimization (PPO), for the task of personalized exam scheduling. We consider several key points, including learning efficiency, the quality of the personalized educational path, adaptability to changes in student performance, scalability with increasing numbers of students and courses, and implementation complexity. Experimental results, based on case studies conducted within a single degree program at a university, demonstrate that, while Q-Learning offers simplicity and greater interpretability, PPO offers superior performance in handling the complex and stochastic nature of students’ learning trajectories. Experimental results, conducted on a dataset of 391 students and 5700 exam records from a single degree program, demonstrate that PPO achieved a 42.0% success rate in improving student scheduling compared to Q-Learning’s 26.3%, with particularly strong performance on problematic students (41.3% vs 18.0% improvement rate). The average delay reduction was 5.5 months per student with PPO versus 3.0 months with Q-Learning, highlighting the critical role of algorithmic design in shaping educational outcomes. This work contributes to the growing field of AI-based instructional support systems and offers practical guidance for the implementation of intelligent tutoring systems. Full article

University course scheduling is a kind of timetable problem and can be mathematically formulated as an integer linear programming problem. Essentially, a university course scheduling problem is an optimization problem that aims at most efficiently minimizing a cost function according to a set of constraints. The huge searching space for the course scheduling problem means a long time will be needed to find the optimal solution. Therefore, some studies have used soft computing approaches to solve course scheduling problems in order to reduce the searching space. However, in order to use soft computing approaches to solve university course scheduling problems, we may need to design algorithms and conduct numerous experiments to achieve maximum efficiency. Thus, in this study, instead of employing soft computing methods, we propose a SWI-PROLOG-based expert system to solve the course scheduling problem. An experiment was conducted using real-world data from a department at a national university in southern Taiwan. During the experiment, each teacher in the department chose five preferential time slots. The experimental results have shown that about 99% of courses were scheduled in teachers’ five preferential time slots with an acceptable computational time of executing SWI-PROLOG (127 milliseconds on a regular personal computer). This study has thus provided a framework for solving course scheduling problems using an expert system. This would be the main contribution of this study. Full article

Decarbonizing production activities is a critical task in the transition towards carbon neutrality. Traditional carbon footprint accounting tools, such as life-cycle assessment (LCA) and the Greenhouse Gas Protocol, primarily quantify direct and indirect emissions but offer limited guidance on actionable reduction strategies. To address this gap, this study proposes a comprehensive life-cycle carbon footprint optimization framework that integrates LCA with a mixed-integer linear programming (MILP) model. The framework, while applicable to various production contexts, is validated using a university campus as a case study. In 2023, the evaluated university’s net carbon emissions totaled approximately 24,175.07 t CO₂-eq. Based on gross emissions (28,306.43 t CO₂-eq) before offsetting, electricity accounted for 66.09%, buildings for 15.55%, fossil fuels for 8.67%, and waste treatment for 8.46%. Seasonal analysis revealed that June and December exhibited the highest energy consumption, with emissions exceeding the monthly average by 19.4% and 48.6%, respectively, due to energy-intensive air conditioning demand. Teaching activities emerged as a primary contributor, with baseline emissions estimated at 5485.24 t CO₂-eq. Optimization strategies targeting course scheduling yielded substantial reductions: photovoltaic-based scheduling reduced electricity emissions by 7.00%, seasonal load shifting achieved a 26.92% reduction, and combining both strategies resulted in the highest reduction, at 45.95%. These results demonstrate that aligning academic schedules with photovoltaic generation and seasonal energy demand can significantly enhance emission reduction outcomes. The proposed framework provides a scalable and transferable approach for integrating time-based and capacity-based carbon optimization strategies across broader operational systems beyond the education sector. Full article

Objective: This study intends to evaluate simulated experiences provided to pharmacy students that directly compare the perspective of patients managing chronic disease states through traditional means without compliance aids to those using compliance aids, such as continuous glucose monitors (CGMs) and other devices. Methods: This simulation was conducted with third-year pharmacy students enrolled in the ambulatory care elective course at the University of Florida College of Pharmacy. It was designed to simulate a patient responsible for self-administering an array of medications for multiple chronic diseases that the students are likely to encounter during clinical practice. For the first week, students were tasked with adhering to a complex medication schedule from their associated pill bottles without the use of compliance aids (pill organizers, alarms, etc.) and checking their blood glucose twice daily using a traditional glucometer. In the second week, students continued the role of the patient; however, they were provided with compliance aids and encouraged to set alarms and use CGMs. Using a questionnaire developed based on the traditional Likert scale model, the students were able to quantify their experiences in a way that allowed the investigators to observe any changes. Results: Regarding the overall implications of this experience, most participants (>80%) agreed that this project increased their understanding of the value of compliance aids and devices and encouraged them to not only incorporate them into their future patient care plans but also advocate for accessibility to improve health outcomes. Conclusion: Students who completed this experience reported better adherence to chronic disease state control using compliance aids and, in turn, the applicability of the use of compliance aids in managing those with complex medication regimens. This simulation may encourage future pharmacists to incorporate compliance aids with their patients to improve health outcomes. Full article

The university course scheduling problem (UCSP) is a challenging combinatorial optimization problem that requires optimization of the quality of the schedule and resource utilization while meeting multiple constraints involving courses, teachers, students, and classrooms. Although various algorithms have been applied to solve the UCSP, most of the existing methods are limited to scheduling independent courses, neglecting the impact of joint courses on the overall scheduling results. To address this limitation, this paper proposed an innovative mixed-integer linear programming model capable of handling the complex constraints of both joint and independent courses simultaneously. To improve the computational efficiency and solution quality, a hybrid method combining a genetic algorithm and dynamic programming, named POGA-DP, was designed. Compared to the traditional algorithms, POGA-DP introduced exchange operations based on a judgment mechanism and mutation operations with a forced repair mechanism to effectively avoid local optima. Additionally, by incorporating a greedy algorithm for classroom allocation, the utilization of classroom resources was further enhanced. To verify the performance of the new method, this study not only tested it on real UCSP instances at Beijing Forestry University but also conducted comparative experiments with several classic algorithms, including a traditional GA, Ant Colony Optimization (ACO), the Producer–Scrounger Method (PSM), and particle swarm optimization (PSO). The results showed that POGA-DP improved the scheduling quality by 46.99% compared to that of the traditional GA and reduced classroom usage by up to 29.27%. Furthermore, POGA-DP increased the classroom utilization by 0.989% compared to that with the traditional GA and demonstrated an outstanding performance in solving joint course scheduling problems. This study also analyzed the stability of the scheduling results, revealing that POGA-DP maintained a high level of consistency in scheduling across adjacent weeks, proving its feasibility and stability in practical applications. In conclusion, POGA-DP outperformed the existing algorithms in the UCSP, making it particularly suitable for efficient scheduling under complex constraints. Full article

Solving the university timetabling problem is crucial as it ensures efficient use of resources, minimises scheduling conflicts, and enhances overall productivity. This paper presents a comprehensive review of university timetabling problems using integer programming algorithms. This study explores various integer programming techniques and their effectiveness in optimising complex scheduling requirements in higher education institutions. We analysed 95 integer programming-based models developed for solving university timetabling problems, covering relevant research from 1990 to 2023. The goal is to provide insights into the evolution of these algorithms and their impact on improving university scheduling. We identify that the implementation rate of models using integer programming is 98%, which is much higher than 34% implementation rates using meta-heuristics algorithms from the existing review. The integer programming models are analysed by the problem types, solutions, tools, and datasets. For three types of timetabling problems including course timetabling, class timetabling, and exam timetabling, we dive deeper into the commercial solvers CPLEX (47), Gurobi (11), Lingo (5), Open Solver (4), C++ GLPK (4), AIMMS (2), GAMS (2), XPRESS (2), CELCAT (1), AMPL (1), and Google OR-Tools CP-SAT (1) and identify that CPLEX is the most frequently used integer programming solver. We explored the uses of machine learning algorithms and the hybrid solutions of combining the integer programming and machine learning algorithms in higher education timetabling solutions. We also identify areas for future work, which includes an emphasis on using integer programming algorithms in other industrial areas, and using machine learning models for university timetabling to allow data-driven solutions. Full article

Background: The COVID-19 pandemic forced medical programs to rapidly switch to remote teaching from scratch, impacting hands-on skills training. This study compared the efficacy of a hybrid online format to a regular in-person session for a mandatory surgical skills class. Methods: Third-year undergraduate medical students attending the surgical skills class in the winter semester of 2020/21 at the Medical University of Graz were randomly assigned to either the hybrid or in-person class, depending on their course schedule and government regulations. The hybrid class involved online videos, one-on-one peer tutoring, and an Objective Structured Clinical Examination (OSCE). Pre- and post-class self-assessments were conducted to evaluate their theoretical and practical knowledge of a single interrupted suture. Results: The study included 85 students in the regular in-person class and 50 in the hybrid class. A pre-class assessment revealed higher self-assessments in the hybrid class for theoretical and practical knowledge, but a post-class assessment showed no significant difference. The advantages and disadvantages of both modalities were identified, providing valuable insights for future curriculum development. Conclusions: Both teaching modes were effective for undergraduate surgical skills training. This study recommends implementing positive aspects of both the hybrid and in-person formats while recognizing their respective limitations. Full article

University course scheduling (UCS) is one of the most important and time-consuming issues that all educational institutions face yearly. Most of the existing techniques to model and solve UCS problems have applied approximate methods, which differ in terms of efficiency, performance, and optimization speed. Accordingly, this research aims to apply an exact optimization method to provide an optimal solution to the course scheduling problem. In other words, in this research, an integer programming model is presented to solve the USC problem. In this model, the constraints include the facilities of classrooms, courses of different levels and compression of students’ curriculum, courses outside the faculty and planning for them, and the limited time allocated to the professors. The objective is to maximize the weighted sum of allocating available times to professors based on their preferences in all periods. To evaluate the presented model’s feasibility, it is implemented using the GAMS software. Finally, the presented model is solved in a larger dimension using a real data set from a college in China and compared with the current program in the same college. The obtained results show that considering the mathematical model’s constraints and objective function, the faculty courses’ timetable is reduced from 4 days a week to 3 working days. Moreover, master courses are planned in two days, and the courses in the educational groups do not interfere with each other. Furthermore, by implementing the proposed model for the real case study, the maximum teaching hours of the professors are significantly reduced. The results demonstrate the efficiency of the proposed model and solution method in terms of optimization speed and solution accuracy. Full article

The emergence of immersive technologies, such as virtual reality (VR) headsets, has revolutionized the way we experience the physical world by creating a virtual, interactive environment. In the field of education, this technology has immense potential to provide students with a safe and controlled environment in which to experience real-world scenarios that may be otherwise unfeasible or unsafe. However, limited research exists on the effectiveness of integrating immersive technologies into technical education delivery. This research investigated the potential use of immersive virtual reality (IVR) in university-level construction management courses, with a focus on integrating IVR technology into traditional education for construction project planning and control. The experiment involved comparing the students’ learning and understanding of the subject matter using a set of two-dimensional construction drawings and a critical path method (CPM)-based construction schedule, with and without the use of an immersive environment. The findings suggested that the use of immersive technology significantly improved the students’ ability to understand technical concepts and identify any errors in the construction sequence when compared to traditional teaching methods. This paper presents the details of the experiment and a comparative analysis of both approaches in terms of students’ learning and understanding of project planning, sequencing, and scheduling. Full article

Knowing the attitudes and behaviors of nursing students in relation to vaccination is important because they will soon be determinant for the health literacy of the population. Vaccination remains the most effective response in the fight against communicable diseases, including COVID-19 and influenza. The objective of this study is to analyze the attitudes and behaviors of Portuguese nursing students with regard to vaccination. A cross-sectional study was carried out, with data collection from nursing students at a university in Lisbon, Portugal. A sample of 216 nursing students was obtained, representing 67.1% of the students enrolled in this university. What stands out from the results of the questionnaire “Attitudes and Behaviors in Relation to Vaccination among Students of Health Sciences” is that for the majority of students the answers were positive; in addition, 84.7% had a completed vaccination schedule for COVID-19. Being a nursing student, being in the final years of the course and being a woman are the factors that most influence the positive attitude of the students. The results obtained are motivating, because these students will be the future health professionals most likely to integrate health promotion programs through vaccination. Full article

With information technology pushing the development of intelligent teaching environments, the online teaching platform emerges timely around the globe, and how to accurately evaluate the effect of the “any-time and anywhere” teacher–student interaction and learning has become one of the hotspots of today’s education research. Bullet chatting in online courses is one of the most important ways of interaction between teachers and students. The feedback from the students can help teachers improve their teaching methods, adjust teaching content, and schedule in time so as to improve the quality of their teaching. How to automatically identify the sentiment polarity in the comment text through deep machine learning has also become a key issue to be automatically processed in online course teaching. The traditional single-layer attention mechanism only enhances certain sentimentally intense words, so we proposed a sentiment analysis method based on a hierarchical attention mechanism that we called HAN. Firstly, we use CNN and LSTM to extract local and global information, gate mechanisms are used for extracting sentiment words, and the hierarchical attention mechanism is then used to weigh the different sentiment features, with the original information added to the attention mechanism concentration to prevent the loss of information. Experiments are conducted on China Universities MOOC and Tencent Classroom comment data sets; both accuracy and F1 are improved compared to the baseline, and the validity of the model is verified. Full article

(1) Background: During the past 3 years, the COVID-19 pandemic has severely affected the normal school schedule of college students, jeopardizing their mental health, sleep quality, and interpersonal relationships. However, previous studies have focused on the dimension of social support received, and few studies have measured in depth the association of support received from family on adolescents’ physical and mental health. Therefore, this study explored the associations between family support received by Chinese college students during COVID-19 pandemic online classes, stress and sleep quality, and the mediating role of stress. (2) Methods: A cross-sectional study conducted at Chongqing Medical University recruited 712 college students through a university-wide incidental random sample using the Questionnaire Star platform. Statistical description and correlation analysis was conducted using SPSS 25.0, and structural equation modeling was constructed using AMOS 22.0 to test for mediating effects; (3) Results: The family support score of college students during the COVID-19 pandemic online course was 19.41 ± 4.62. Correlation analysis showed that sleep quality was negatively correlated with family support (r = −0.224, p < 0.01), positively correlated with stress (r = 0.324, p < 0.01), and family support was negatively correlated with stress (r = −0.159, p < 0.01). The results of structural equation modeling showed that stress partially mediated the relationship between family support and sleep quality among college students (indirect effect = −0.150, p < 0.01, SE = 0.013,95% CI = [−0.208, −0.064]). The model R2 was 36.4%. (4) Conclusions: Schools should consider implementing sleep education, and stress relief curriculum measures to improve the quality of students’ sleep, and should focus on the role that family plays during online classes. This will help students overcome the negative emotional effects of stress in the COVID-19 pandemic and improve their learning efficiency and physical and mental health. Full article

Accreditation is widely considered to be a vital tool for quality assurance in higher education; however, there is disagreement in the academic community on the intended benefits of accreditation. Preparing for accreditation requires extensive financial and human resources to complete the required documentation. All accreditation agencies require improvements in institutional infrastructure, enhanced student support, appropriate learning environments, and faculty development, which can directly improve students’ learning experiences. In this paper, we explore the impact of accreditation on students’ learning by using a case study-based approach. We selected four degree programs from a University in Saudi Arabia and compared the performances of students in different courses before and after acquiring local program accreditation (NCAAA). The results highlight that although there is no direct relationship between increased student performance and acquiring accreditation, there is a significant impact on the performance of student learning. However, there is a need for sustained efforts to continuously adopt accreditation-aligned practices to gain a sustained advantage. We have presented a model that can enable academic institutions to continuously adhere to best practices even if no accreditation visit has been scheduled in the near future. This way, academic programs can consistently improve their processes and enhance student learning. Full article

The occurrence and course of immune-mediated diseases (IMDs) following COVID-19 vaccination has been little explored so far. We retrieved, among adult patients hospitalized at the Internal Department of a French university hospital up to May 2022, all those who had developed, or relapsed to, an IMD less than 3 weeks following COVID-19 vaccination, without other triggers. Twenty-seven (24 new-onset) post-COVID-19 vaccine IMDs were recorded. They comprised giant cell arteritis or polymyalgia rheumatica (n = 16, HLA-DRB1*04 in 58% of 12 assessed GCA cases), immune-mediated necrotizing myositis or acute rhabdomyolysis, systemic vasculitis, immune thrombocytopenic purpura, rheumatoid arthritis, anti-synthetase syndrome, and adult-onset Still’s disease. The causative vaccines were mRNA-based (20 cases) or viral vector-based (7 cases). The IMD typically occurred after the first vaccine dose, with an average delay of 8 (5 SD) days. The patients’ mean age was 67 years, and 58% were women. The IMDs had protracted courses in all but three of the patients and typically required high-dose glucocorticoids, in combination with immunomodulators in 13 patients. One patient died of intractable rhabdomyolysis, whereas five suffered permanent damage from IMDs. Eleven patients with well-controlled IMDs completed their COVID-19 vaccination schedule, and two suffered mild IMD relapses. There is a risk of IMDs, notably GCA/PMR, and muscle disorders, following COVID-19 vaccination. Such adverse reactions typically occurred after the first dose, raising concern about subsequent COVID-19 vaccinations. However, early re-challenge in well-controlled IMDs appeared safe. Full article