Learning Preferences Research Articles

Preference learning based on adaptive graph neural network for multi-criteria decision support

Intelligent decision-making assists decision-makers (DMs) in making choices through data analysis, model prediction, and automated processes. Central to this field are two key concepts: multi-criteria decision making (MCDM) and preference learning (PL). While MCDM and PL both aim to develop decision models that rank alternatives based on observed or revealed preferences, they diverge in focus. MCDM concentrates on the DMs' perspectives, whereas PL emphasizes model-driven approaches. This divergence presents significant challenges in integrating these methodologies, particularly in ensuring the integrated method remains scalable and interpretable amidst the complexity of decision scenarios. To bridge this gap, our study introduces the use of graph structures to frame decision problems and proposes a novel PL method employing graph neural network (GNN) for multi-criteria decision support. This method is anchored in the Technique for Order Preference by Similarity to an Ideal Solution (TOPSIS) technique and combines an adaptive GNN model with a weight determination model. The GNN model updates embeddings from the alternative's criteria and category features, utilizing an attention mechanism to adaptively assess their importance. Concurrently, the weight determination model contains a weight neural network module to set objective criterion weights and a game theory-based module for calculating combined criterion weights. The method not only inherits the interpretability and intuitive appeal of decision models but also leverages the computational efficiency and high accuracy of machine learning. In experiments conducted on benchmark datasets, our method exhibits significant performance improvements, especially in ranking-related evaluation metrics, outperforming the best baseline by 5.78%.

Personalized learning model based on machine learning algorithms

<p>Machine learning algorithms have been widely applied in the field of personalized learning within educational information technology. By leveraging big data analysis and data mining techniques, machine learning can help identify patterns and trends in students' learning behaviors, preferences, and performance. This information can then be used to tailor educational resources and experiences to meet the individual needs and unique characteristics of each learner. Machine learning has made great progress and achievements in the teaching process of universities, but there are also some shortcomings. Such as data dependence, over-fitting and under-fitting, explanatory problems, need a lot of computing resources, data bias, sensitive to outliers, cannot solve all problems, and the challenge of data privacy, through the analysis of machine learning algorithm model, efforts to find ways to expand the dimension of personalized learning classroom, meet the students in learning objectives, learning content, learning methods of the special characteristics and unique needs, to guide students to actively explore and research, obtain innovation and appropriate learning results.</p>

Machine learning-based techno-econo-environmental analysis of CO2-to-olefins process for screening the optimal catalyst and hydrogen color

CO2 to light olefins (CTLO) is an attractive strategy for CO2 fixation to obtain high-value-added chemical products. Selecting the optimal catalyst and hydrogen source for the process has become increasingly challenging due to the wide range of developed catalysts and diverse hydrogen sources available. Herein, this study proposes a machine learning-based techno-econo-environmental analysis framework for screening catalysts and hydrogen colors for the CTLO process. The preferred machine learning model is employed to screen the most promising catalyst for the CTLO process, which is then applied to accomplish the conceptual design of the entire process to obtain essential material and energy balance results by process modeling and simulation. Finally, the techno-econo-environmental performance of the CTLO processes using various colored hydrogen are compared to provide more informed choices. The results indicate the extreme gradient boosting model exhibits superior predictive accuracy, making it suitable for integrating with the genetic algorithm to screen the optimal catalyst for the CTLO process. After multi-objective optimization, an optimal Fe-based catalyst is screened and used for the modeling and simulation of the CTLO process due to its highest light olefins yield (36.43%). The carbon, hydrogen, and exergy utilization efficiencies of the CTLO process considering solely the target light olefins are 59.58%, 21.71%, and 40.41%, respectively. After evaluating the impact of different H2 colors on the total operating cost, unit production cost, and minimum selling price of the CTLO process, it was found the CTLO process using gray H2 generated by coke oven gas and coal gasification technologies exhibits the most favorable cost-effectiveness and optimal market price compared to alternative production scenarios. However, using green H2 proves highly advantageous for the CTLO process in attaining comprehensive negative carbon emissions throughout its lifecycle. Moreover, it anticipates the influence of technological advancements and market fluctuations on the market competitiveness of CTLO processes with diverse hydrogen colors.

Revisiting Long- and Short-Term Preference Learning for Next POI Recommendation With Hierarchical LSTM

Revisiting Long- and Short-Term Preference Learning for Next POI Recommendation With Hierarchical LSTM

Machine Learning Algorithm as a Firewall Decision and Reinforcement in Market Segmentation and Big Data

In the modern digital landscape, vast amounts of data are generated daily, challenging traditional analytical approaches in both data security and market segmentation. This paper presents a machine learning (ML) algorithm designed to serve a dual purpose: as a decision-support firewall system and as a reinforcement tool for market segmentation within big data environments. By integrating supervised and unsupervised learning models, the proposed algorithm effectively detects anomalies and potential security threats while also identifying distinct customer segments with high precision. The firewall decision-making component utilizes predictive models to detect malicious activity in real-time, enhancing cybersecurity by proactively learning from previous threat patterns. Concurrently, the reinforcement learning component analyses customer behaviour and preferences, dynamically adapting segmentation strategies to maximize marketing effectiveness. The dual implementation of ML in these domains demonstrates significant potential in improving both data security and personalized marketing outcomes. Experimental results indicate enhanced firewall accuracy and a refined segmentation process, suggesting that the proposed ML model provides a comprehensive solution for the challenges posed by big data in cybersecurity and market segmentation. s. Keywords Machine Learning; Firewall Decision; Reinforcement Learning; Market Segmentation; Big Data

Introduction Strategies for English Extracurricular Masterpieces

Comprehension Passages: A Guide for English Teachers provides a comprehensive overview of strategies and methodologies to enhance reading comprehension among English language learners. It begins by emphasising the crucial role of reading comprehension in language learning and identifying the common challenges students face while dealing with comprehension passages. The article outlines the cognitive processes involved in reading comprehension, including the decoding of text and the integration of new information with existing knowledge, and discusses the common obstacles students encounter. Key teaching strategies for reading comprehension are explored in detail. These include pre - reading strategies that activate prior knowledge and set purposes for reading, during - reading strategies that maintain engagement and foster understanding, and post - reading strategies that reinforce learning and evaluate comprehension. The integration of technology and multimedia in reading lessons is highlighted, emphasising the use of digital tools and interactive content to cater to diverse learning styles and preferences. Additionally, the article discusses the importance of adapting teaching methods to accommodate various learners, underscoring the need for understanding different learning styles in the context of reading comprehension. It also delves into the role of assessment and feedback in reading education, outlining effective methods for assessing reading comprehension skills and providing constructive feedback to improve student performance. Overall, the article serves as a practical guide for English teachers, offering insights and strategies to effectively navigate the complexities of teaching reading comprehension. It aims to equip educators with the tools necessary to enhance their students' reading skills, thereby fostering a deeper understanding and appreciation of the written word.

AI Enable Personalized Assistive Tools to Enhance Education of Disabled Persons and Its Legal Fallacies

The integration of AI-enabled personalized assistive tools in education represents a transformative approach to enhancing learning experiences for disabled persons and children with diverse needs. This abstract explores the potential of these technologies to provide tailored educational support, facilitating engagement and fostering independence. By leveraging machine learning algorithms, natural language processing, and adaptive learning environments, these tools can customize content delivery based on individual learning styles, preferences, and abilities. AI-driven platforms can assist in identifying specific challenges faced by students, offering real-time feedback and adaptive resources that evolve with their progress. Examples include speech recognition software for students with hearing impairments, interactive visual aids for those with cognitive disabilities, and personalized learning pathways that accommodate varying levels of ability. Furthermore, the implementation of these tools promotes inclusivity in educational settings, allowing all learners to participate meaningfully. This abstract highlight the need for collaboration between educators, technologists, and policymakers to ensure the effective design and deployment of AI assistive technologies. By prioritizing accessibility and user-centric design, these innovations can significantly enhance educational outcomes, empower disabled individuals, and cultivate a more equitable learning environment.

Bumblebee social learning outcomes correlate with their flower-facing behaviour

Previous studies suggest that social learning in bumblebees can occur through second-order conditioning, with conspecifics functioning as first-order reinforcers. However, the behavioural mechanisms underlying bumblebees’ acquisition of socially learned associations remain largely unexplored. Investigating these mechanisms requires detailed quantification and analysis of the observation process. Here we designed a new 2D paradigm suitable for simple top-down high-speed video recording and analysed bumblebees’ observational learning process using a deep-learning-based pose-estimation framework. Two groups of bumblebees observed live conspecifics foraging from either blue or yellow flowers during a single foraging bout, and were subsequently tested for their socially learned colour preferences. Both groups successfully learned the colour indicated by the demonstrators and spent more time facing rewarding flowers—whether occupied by demonstrators or not—compared to non-rewarding flowers. While both groups showed a negative correlation between time spent facing non-rewarding flowers and learning outcomes, the observer bees in the blue group benefited from time spent facing occupied rewarding flowers, whereas the yellow group showed that time facing unoccupied rewarding flowers by the observer bees positively correlated with their learning outcomes. These results suggest that socially influenced colour preferences are shaped by the interplay of different types of observations rather than merely by observing a conspecific at a single colour. Together, these findings provide direct evidence of the dynamical viewing process of observer bees during social observation, opening up new opportunities for exploring the details of more complex social learning in bumblebees and other insects.

Parental awareness and perspectives on newborn screening in China: a questionnaire-based study

BackgroundLow parental awareness and knowledge about newborn screening have been identified as a public issue. This study explored Chinese parents’ self-evaluation of awareness, knowledge, and methods of receiving information about newborn screening.MethodsUsing convenience sampling, we included 614 respondents who were expectant parents or parents of children aged 0-3 years. Our self-made questionnaire comprised four sections: sociodemographic characteristics, self-evaluation of awareness, detailed knowledge about newborn screening, and practical and expected methods of receiving newborn screening information.ResultsWe found that 72.9% of participants were classified as aware of newborn screening. However, only 14.2% of the participants received a passing score on the newborn screening detailed knowledge questions. Knowledge level about newborn screening was significantly associated with gender (P < .001), age (P < .05), education level (P < .05), residence (P < .05), family income (P < .05), and number of children (P < .05). The knowledge acquisition about newborn screening mainly came from hospital-related training (62.1%). Additionally, nearly half of the respondents (48.0%) expressed a preference for learning more about newborn screening through social media platforms, such as WeChat.ConclusionsWhile the majority of expectant or new parents were aware of newborn screening, only a minority had a thorough understanding of it. Various sociodemographic factors were associated with the level of parental knowledge about newborn screening. It is recommended to use hospital lectures or social media initiatives to educate parents in China.

Analysis of Learning Style Profiles to Optimize Student Achievement

Student achievement is an indicator of learning success that indicates the extent to which a student has acquired the skills and knowledge set forth by their educational program. Understanding students' learning styles plays a significant role in determining learning success. One of the efforts to optimize student achievement is facilitating learning activities that align with students' learning styles. Students are no longer preoccupied with understanding their learning styles but are focused on deepening the material. This study aims to describe the learning style profiles of students, and students and lecturers in the learning process can follow up on the results. This study uses a survey research method with a questionnaire instrument containing 54 statements. A learning styles questionnaire prepared based on the opinion of De Poter &amp; Hernacki is used as a data collection tool. Learning styles are identified as visual learning style, auditorial learning style, and kinesthetic learning style. The questionnaire was distributed to students in semesters 2, 4, 6, and 8, totaling 51 students. The data analysis technique used is a descriptive analysis of the questionnaire results. The questionnaire results show that 36% of students have visual and kinesthetic learning styles, 14% have auditory, and 14% are inconsistent. This inconsistency may suggest a balanced learning style, where students do not rely heavily on one learning mode but rather integrate multiple styles depending on the context. Therefore, the Science Education Study Program lecturers should implement a learning model that can facilitate visual and kinesthetic styles, even though the subjects taught are theoretical. Incorporating a variety of teaching methods, such as visual aids, hands-on activities, and auditory discussions, can accommodate the diverse learning preferences of students, leading to a more inclusive and effective learning environment.

Factors Influencing Students' Learning Preferences in College Piano Teaching and Countermeasures

In the course of higher education development, piano teaching plays a vital role in fostering students' basic musical literacy, guiding them to form good music learning habits, helping them establish a fundamental understanding of music knowledge, and thereby enhancing their music appreciation abilities. Due to the diversification of educational environments and the increasing personalization of student demands, in order to improve the quality of piano music teaching and stimulate students' potential, it is crucial to study students' learning preferences from various angles. This forms an important foundation for personalized teaching. Therefore, university piano teaching still needs continuous improvement in teaching strategies. Piano teaching should be adjusted according to course requirements, abandoning outdated piano education concepts and rigid methods, and analyzing from multiple dimensions to clarify the connection between students' learning preferences and the effectiveness of piano teaching. This helps students master piano playing techniques and strengthen their skills, equipping them with solid basics and advancing college piano education to new heights.

Moving beyond tokenism: Sustaining engagement of persons living with dementia in identifying emergency research priorities.

The Geriatric Emergency Care Applied Research Network 2.0-Advancing Dementia Care (GEAR 2.0-ADC) aims to advance research efforts to improve the emergency care of persons living with dementia (PLWDs). To support this objective, GEAR 2.0-ADC convened a virtual consensus conference to prioritize emergency care research opportunities for PLWDs inclusive of perspectives of PLWDs to ensure identification of research gaps in response to their experiences and priorities. Inclusion of PLWDs as research partners is increasingly recognized as a best practice, however, approaches to facilitating consensus participation are lacking. Best practices for supporting the engagement of PLWDs in a consensus conference, applied across its three phases (pre-conference, during the conference, and post-conference), include: establishing a learning environment focused on research priorities before the event, presenting information in ways that align with participants' learning preferences while accommodating cognitive impairments, and providing multiple opportunities and methods for gathering post-conference feedback from PLWDs. These strategies were identified by PLWDs and care partners (CPs) through semi-structured interviews, who were involved in the convening process, aimed at exploring ways to enhance facilitation techniques for participants. Additionally, these summarized insights aim to encourage the use of community-engaged approaches in discussions and consensus-building around research priorities in emergency care, particularly for PLWDs and their CPs.

Enhancing Academic Performance with AI-Powered Tools: A Comparative Study of Adobe Photoshop and Lightroom in Educational Technology Photography

Artificial Intelligence (AI) encompasses various technologies that enable machines to perform tasks that typically require human intelligence, including educational applications with photography not left behind. This study investigates the comparative effects of AI tools in Adobe Photoshop and Lightroom on undergraduates’ academic performance in an Educational Technology Photography course university of Ilorin. A total of 42 students participated, with 52.4% trained in Photoshop and 47.6% in Lightroom. The gender distribution was 59.5% male and 40.5% female. Survey results indicated a universally positive perception of both AI tools, with no disagreements recorded. Respondents rated both tools highly in terms of ease of use, preference for learning, confidence in skills, and essentiality for professional editing. Achievement test scores revealed excellent performance (µ &gt; 80%) for students using both tools, with no significant gender differences. The findings suggest that AI-powered tools in Adobe Photoshop and Lightroom significantly enhance students’ learning experiences and academic performance. The study highlights the potential for integrating AI tools in educational settings to improve learning outcomes and recommends further research into their long-term impacts, ethical implications, and effectiveness across diverse disciplines and educational contexts.

What makes students tick? Exploring factors that affect learner motivations and challenges when engaging with optional library workshops on data literacy

Optional data literacy library workshops often struggle to achieve consistent high enrollment and attendance. In seeking strategies to mitigate these challenges, this qualitative study attempted to gain a better understanding of learner motivations and challenges when enrolling and attending optional data literacy instruction offered by a data library at a large university. Nineteen graduate students were interviewed. This study found four factors involved in motivating graduate students to enroll and attend optional library data literacy workshops. These factors, in the form of questions that graduate students would ask about the workshops, were: does the workshop address my need; does the workshop fit into my schedule; does the workshop satisfy my learning preferences; and are there health, emotional, or social aspects that could affect my workshop attendance. Based on these findings, several strategies to improve motivation and attendance were suggested, such as providing workshops at time of need, holding regular repeated sessions, presenting a mix of online and in-person offerings, and making workshops part of a social or networking event.

Dysregulation of neuropilin-2 expression in inhibitory neurons impairs hippocampal circuit development and enhances risk for autism-related behaviors and seizures.

Dysregulation of development, migration, and function of interneurons, collectively termed interneuronopathies, have been proposed as a shared mechanism for autism spectrum disorders (ASDs) and childhood epilepsy. Neuropilin-2 (Nrp2), a candidate ASD gene, is a critical regulator of interneuron migration from the median ganglionic eminence (MGE) to the pallium, including the hippocampus. While clinical studies have identified Nrp2 polymorphisms in patients with ASD, whether selective dysregulation of Nrp2-dependent interneuron migration contributes to pathogenesis of ASD and enhances the risk for seizures has not been evaluated. We tested the hypothesis that the lack of Nrp2 in MGE-derived interneuron precursors disrupts the excitation/inhibition balance in hippocampal circuits, thus predisposing the network to seizures and behavioral patterns associated with ASD. Embryonic deletion of Nrp2 during the developmental period for migration of MGE derived interneuron precursors (iCKO) significantly reduced parvalbumin, neuropeptide Y, and somatostatin positive neurons in the hippocampal CA1. Consequently, when compared to controls, the frequency of inhibitory synaptic currents in CA1 pyramidal cells was reduced while frequency of excitatory synaptic currents was increased in iCKO mice. Although passive and active membrane properties of CA1 pyramidal cells were unchanged, iCKO mice showed enhanced susceptibility to chemically evoked seizures. Moreover, iCKO mice exhibited selective behavioral deficits in both preference for social novelty and goal-directed learning, which are consistent with ASD-like phenotype. Together, our findings show that disruption of developmental Nrp2 regulation of interneuron circuit establishment, produces ASD-like behaviors and enhanced risk for epilepsy. These results support the developmental interneuronopathy hypothesis of ASD epilepsy comorbidity.

Influence of Preferences on the Use of Information Resources by the Library and Information Science Undergraduate Students in South-South, Nigeria

Background: The use of information resources by the Library and Information Science (LIS) undergraduate students in Delta State, Nigeria, was addressed based on diverse learning preferences, a critical factor potentially influencing academic success. Purpose: This research aimed to explore learning preferences among LIS undergraduate students and assess the influence on the use of information resources. Method: A correlational research design was applied, including 325 students selected through multi-stage sampling from seven state-owned universities in Delta, Edo, and Rivers states. Furthermore, data were collected through questionnaires, with 268 valid responses at an 82% retrieval rate. These were analyzed with descriptive statistics (frequency, percentage, and mean) as well as inferential statistics (Pearson’s product-moment coefficient, and ANOVA), with a criterion mean of 2.50 and an alpha level of 0.05. Result: The results showed that the students had varied learning preferences, including verbal presentations (68%), physical interaction with materials (66%), and watching educational videos (64%). Preferences significantly influenced the use of information resources (Agg. x̄ = 2.83), with a positive correlation between these two variables (r = .907, p &lt; 0.05). Additionally, a significant difference was observed between the education level of students and learning preferences (F = 2,265 = 1189.172, p &lt; 0.05). Conclusion: The students were found to benefit majorly from the combination of traditional and modern teaching methods. Mutual collaboration between university librarians and coordinators of information resources with the faculty was recommended to ensure the correspondence of educational materials to learning preferences.

Towards Better Numeracy Outcomes: Analyzing Factors and Interventions for Indonesian Students

Numeracy is essential in many aspects of daily life, yet it remains a growing concern among Indonesian students. Numerous factors contribute to this issue, which must be addressed through pedagogical and psychological approaches. Correspondingly, this research had two main objectives: First, to describe students' performance in basic numeracy skills and the factors that hindered their development, and second, to analyze and map the learning activities students needed to improve their numeracy skills. To achieve these objectives, an ex post facto research design with a quantitative approach was employed. A total of 506 secondary school students from various regions in Indonesia (Female = 285; Male = 221) participated in this study. The sample was collected using a simple random sampling technique. Data was gathered via an online survey of 51 structured questions administered to the students. Multiple statistical tests were conducted to analyze the data, including descriptive statistics (mean scores) and inferential statistics (factor analysis). The findings indicated that the overall mean score for students' basic numeracy skills was moderate, with female students outperforming male students. Additionally, several factors hindered students' mastery of basic numeracy. The study also highlighted the need for more differentiated instruction, where students could receive personalized support based on their unique challenges and learning preferences.

An analysis of factors influencing undergraduate students’ learning preference in nursing laboratory courses with virtual reality technology: A scoping review

An analysis of factors influencing undergraduate students’ learning preference in nursing laboratory courses with virtual reality technology: A scoping review

An Investigation of University Students’ Perceptions of Learning Management Systems: Insights for Enhancing Usability and Engagement

Learning management systems (LMS) have become central to modern education, enabling accessible, personalized, and engaging learning experiences. This study aims to investigate Saudi university students’ perception of LMS in order to explore the critical factors that shape their engagement, satisfaction, and acceptance of these platforms. Drawing from the existing literature that points out the usability challenges of LMS, this study hopes to derive actionable insights to optimize e-learning outcomes. Using Kelly’s repertory grid analysis technique, this study systematically captured and analyzed the personal constructs students associate with LMS, focusing on ease of use, interactivity, and content alignment with educational needs. A sample of 20 university students provided insights on their experiences with LMS features related to usability, functionality, and interactivity, which are critical to engagement. Findings indicate that ease of use is a major determinant of acceptance, along with interactivity and relevant content delivery that supports diverse learning preferences. The study identifies key elements to improve LMS platforms, fostering a more engaging digital learning environment and supporting students’ learning needs. The findings highlight the key aspects: usability of LMS and students’ satisfaction through user-friendly interfaces and interactive features. Institutions that incorporate student feedback into LMS development will likely see improved e-learning outcomes. This research contributes to a deeper understanding of LMS user perceptions and implies refinements that can align platforms with pedagogical demands in higher education.

Differentiated Learning Models: Students’ Learning Style, Readiness, and Interest in Learning Regarding the Results of Fiqh

&lt;em&gt;The differentiation learning model is a novel approach that caters to the demands of the 21st- century education system. Studies conducted by Indonesian scholars on this model are scarce, with limited focus on its impact on students' cognitive abilities and metacognitive awareness. Therefore, this study seeks to investigate the impact of incorporating a differentiated learning approach founded on the learning preferences, preparedness, and engagement of pupils on Fiqh's academic performance. An explanatory survey technique employing the Stratified Random Sampling methodology is utilized in this research. The sample comprises all fifth-grade students of the State Islamic Primary School in Tulungagung Regency, amounting to a total of 507 learners. The study's sample comprises four randomly selected State Islamic Primary Schools from the seven Tulungagung Regency, with a total of 276 students studying Fiqh. The data collection instruments used include test sheets for learning outcomes, notes on creative thinking skills, and motivation questionnaires. The data is then analyzed using a two-way Analysis of Variance (ANOVA) with a significance level of 5%. The results of the research indicate that: (1) There is no discernible disparity in Fiqh learning outcomes based on learning style. (2) There is no disparity in Fiqh learning outcomes based on students’ learning readiness. (3) Disparities in Fiqh learning outcomes based on students’ learning interests exist. (4) A noteworthy interaction exists between the implementation of differentiation learning aspects of learning styles, learning readiness, and students’ learning interests on Fiqh learning outcomes. The findings of this study suggest that the differentiation learning model provides an alternative approach to learning that addresses the needs of the modern era.&lt;/em&gt;