Learning Strategies Research Articles

A solution for the automatic detection of expansion joints in dam stilling pools using underwater robots

This research addresses the critical challenges of detecting and measuring underwater concrete structure expansion joints (CSEJ), with a particular focus on dam stilling basin environments. A full set of algorithms for detecting expansion joints with underwater robots is presented by the research, enabling real-time and offline analysis along with visualization. These algorithms introduce the use of Convolutional Neural Networks (CNNs) for image feature extraction and segmentation in CSEJ detection, significantly improving accuracy and adaptability over traditional methods. An automated approach known as the Underwater Dam Crack Detection Network (UDCD Network) was introduced by this research. And advanced attention gates and hybrid dilated convolutions to enhance its detection capabilities are incorporated in this system. Additionally, a three-phase deep transfer learning strategy during training was utilized in this research, which significantly improves the model's performance. The research suggests an underwater image restoration algorithm that uses a dark channel prior and dynamic white balance correction to improve image clarity. Furthermore, unique underwater environmental challenges were tackled by this research, such as non-contact measurement of expansion joint widths using a view-geometry-based algorithm and comprehensive detection of dam wall surfaces. Experimental validation demonstrates that the advanced imaging technologies of the proposed underwater robots hold significant potential for autonomous detection of expansion joints in dam stilling basins.

A Low-Measurement-Cost-Based Multi-Strategy Hyperspectral Image Classification Scheme.

The cost of hyperspectral image (HSI) classification primarily stems from the annotation of image pixels. In real-world classification scenarios, the measurement and annotation process is both time-consuming and labor-intensive. Therefore, reducing the number of labeled pixels while maintaining classification accuracy is a key research focus in HSI classification. This paper introduces a multi-strategy triple network classifier (MSTNC) to address the issue of limited labeled data in HSI classification by improving learning strategies. First, we use the contrast learning strategy to design a lightweight triple network classifier (TNC) with low sample dependence. Due to the construction of triple sample pairs, the number of labeled samples can be increased, which is beneficial for extracting intra-class and inter-class features of pixels. Second, an active learning strategy is used to label the most valuable pixels, improving the quality of the labeled data. To address the difficulty of sampling effectively under extremely limited labeling budgets, we propose a new feature-mixed active learning (FMAL) method to query valuable samples. Fine-tuning is then used to help the MSTNC learn a more comprehensive feature distribution, reducing the model's dependence on accuracy when querying samples. Therefore, the sample quality is improved. Finally, we propose an innovative dual-threshold pseudo-active learning (DSPAL) strategy, filtering out pseudo-label samples with both high confidence and uncertainty. Extending the training set without increasing the labeling cost further improves the classification accuracy of the model. Extensive experiments are conducted on three benchmark HSI datasets. Across various labeling ratios, the MSTNC outperforms several state-of-the-art methods. In particular, under extreme small-sample conditions (five samples per class), the overall accuracy reaches 82.97% (IP), 87.94% (PU), and 86.57% (WHU).

대학생의 학습양식 기반 학습전략 향상 모바일 앱 프로토타입 개발

Objectives This study aimed to develop a prototype of a mobile application to support the improvement of learning strategies for college students based on their learning styles. Methods With reference to a literature review of learning strategies and styles, strategies were extracted to inform the design of a mobile application aimed at supporting the use of learning strategies based on learning styles. Experts validated the extracted strategies, and the prototype was initially developed accordingly. A usability test was performed with students from University A, and based on the test results, the initial prototype was revised until an evolved version was finalized and presented. Results The strategies were extracted and divided into groups considering such factors as the characteristics of college student users, the connection between learning strategies and academic performance, and the characteristics of mobile applications, respectively. Expert validation of the design strategies indicated that both CVI and IRA were 1.00, securing validity and reliability. The mobile app prototype developed based on the extracted strategies featured menus such as learning style diagnosis, learning strategy recommendation, learning strategy application, learning management, sharing for collaborative learning, and gamification for facilitating learning. The usability test results for the prototype showed that the mean values for all items were 4.25 or higher, signifying positive results. Conclusions The significance of the mobile app prototype presented in this paper lies in its consideration of users’ learning styles, which are among the relevant user characteristics, to recommend appropriate learning strategies and support users as they apply those learning strategies in reality. The study results indicate that utilizing learning styles and gamification could positively influence college students by motivating them to enhance their understanding of learning strategies while also improving their ability to utilize those learning strategies.

Analysis of English Learning Strategies Performed by Successful EFL Learners in Non-English Department

The objectives of this research were to know and describe English learning strategies that are used by successful EFL learners in non-English departments at the University of Mataram and the factors driving them. This is a descriptive qualitative study using purposive sampling. The samples were 5 Engineering students and 5 Economics students with TOEFL scores of 457 to 560. The data was collected through the Strategy Inventory for Language Learning (SILL) questionnaire version 7.0 and semi-interview. The result of the study reveals that successful EFL learners in Engineering and Economics implement all learning strategies at two different levels. The learners use compensation, metacognitive, and cognitive strategies at a high level, then memory, social, and affective strategies at a medium level. This research found that each learning strategy that the learners performed is intertwined. The naturalistic factors that contribute to driving the appropriation and implementation of those strategies are their English proficiency, learning style, and motivation. While there is no significant influence from gender identity found in this research.

Learning Strategies for Chinese as Foreign Language Learners in College: A Qualitative Study

Embarking on the journey of learning Chinese as a foreign language in college entails grappling with a myriad of distinctive challenges, ranging from navigating its intricate writing system to mastering its tonal pronunciation and unraveling its rich cultural nuances. In this complex landscape of language acquisition, comprehending the nuanced learning strategies wielded by Chinese foreign language learners becomes paramount. These strategies not only illuminate the pathways to effective teaching but also hold the key to enhancing student outcomes. Despite the burgeoning interest in Chinese language education, there exists a conspicuous dearth of qualitative studies delving into the intricacies of learning strategies within college settings. To address this gap, this study adopts a qualitative research paradigm, employing semi-structured interviews as its methodological backbone. Through these interviews, the study endeavors to excavate profound insights into the diverse repertoire of learning strategies embraced by Chinese foreign language learners in the collegiate milieu. At its core, the study's primary objective is twofold: to meticulously explore and rigorously analyze the spectrum of learning strategies woven into the fabric of Chinese foreign language learning at the college level. By peering into the multifaceted approaches employed by students, the study aspires to engender a nuanced comprehension of effective pedagogical methodologies. Moreover, the findings are poised to serve as a compass, guiding the design and implementation of tailored language learning interventions tailored to the unique needs and aspirations of Chinese language learners in college environments.

Innovative Strategies for English Language Learning in Ecuador: A Focus on Playful Learning

Education is a universal right that extends beyond core subjects to include cross-curricular content essential for preparing students for a competitive global landscape. In this context, English serves as a critical tool, recognized as both the first official language for 370 million users and a second language for 978 million (Eberhard et al., 2021). Despite its global importance, Ecuador faces challenges in English language proficiency, ranking low in Latin America due to inadequate methodologies and a lack of qualified teachers (Education First, 2021; Banegas & Sanchez, 2024). This project aims to implement a playful program tailored to middle school students (5th to 7th grades) at Eugenio Espejo Basic Education School, addressing their unique educational needs through the development of didactic materials and training for non-specialized teachers. A descriptive quantitative approach will guide the methodology, focusing on communicative competence through structured tutoring sessions that utilize reusable educational resources. The study will assess students' prior knowledge through a pre-test, which will inform the creation of a tailored curriculum designed to enhance vocabulary and overall language skills. This initiative not only seeks to improve English proficiency among students but also aims to equip teachers with effective pedagogical strategies, fostering a more conducive learning environment in Ecuadorian classrooms.

Trainee Teachers’ Shift towards Sustainable Actions in Their Daily Routine

The argument made in this study is that sustainability is a system of beliefs that extends beyond policy and the classroom to transform actions and societies into pro-sustainable behaviours. Therefore, it is crucial to equip trainee teachers to embrace their role as catalysts for change in driving Sustainable development goals, sustainable thinking, systems thinking, wise consumption, sustainable competencies, and sustainable action in their daily lives. The lack of practice-led research on incorporating Education for Sustainable Development (ESD) into trainee teachers’ training programmes is noted in the literature. This qualitative practice-led research explored trainee mathematics and technology teachers’ (TMTTs) experiences learning about sustainability. Kolb’s experiential theory framed this study theoretically. The data were collated from twenty purposively selected TMTTs via semi-structured interviews and reflective diaries. All ethical protocols were observed. The analysis highlighted the core experiences TMTTs gained in learning about sustainability, transformative learning, design thinking, wise consumption, agency, and sustainable teaching strategies. Theoretically, the findings emphasise the importance of using an ESD perspective and experiential learning methods to integrate sustainability education into mathematics and technology teaching. The results suggest that when TMTTs are encouraged to actively engage with sustainability issues in their communities, informed decisions are made about their future roles as teachers, the teaching strategies they intend to use, and the type of learning they aim to foster in their learners.

머신러닝과 HLM을 적용한 중학생의 자기조절 학습전략 중 행동조절에 대한 학생/부모와 교사/학교 수준의 영향요인 검증

Objectives This study focuses on behavioral regulation, the most effectively teachable aspect of self-regulated learning strategies, with the aim of exploratively identifying the factors that influence behavioral regulation in first-year middle school students. Specifically, using multilevel analysis, the study examines not only individual characteristics and parental factors but also teacher and school characteristics, in order to explore potential intervention points at different levels within the home and school environments. Methods Using data from the first year of the Seoul Educational Longitudinal Study of Students 2020, excluding missing values, a total of 4,615 middle school students from 98 schools were analyzed, along with 206 student/parent-level and 188 teacher/school-level explanatory variables selected through data pre-processing. Machine learning techniques, capable of handling multiple variables simultaneously in the model, were employed to identify algorithms with excellent performance in predicting self-regulated learning attitudes at the student/parent level (Level 1) and the teacher/school level (Level 2). The key variables for predicting self-regulated learning attitudes identified at each level were then verified using a two-level hierarchical linear modeling (HLM). Results First, based on the top three algorithms at each level, 10 key variables were derived at the student/parent level and 5 key variables at the teacher/school level. Second, the application of the two-level HLM revealed that learning methods and efforts, growth mindset, goal orientation, human rights, support from academic and career programs, and cognitive empathy at the student/parent level, and the proportion of multicultural students at the teacher/school level, significantly influenced self-regulated learning strategies. Conclusions Based on these results, specific strategies to enhance self-regulated learning attitudes among middle school students were discussed.

LTFM: Long-tail few-shot module with loose coupling strategy for mineral spectral identification

In recent years, deep learning methods have exhibited superior performance in mineral identification when especially compared with conventional machine learning methods such as Support Vector Machine (SVM) and Partial Least Squares (PLS). Nevertheless, almost all of these deep learning methods pay more attention to improving and designing network structures, while neglecting the phenomenon of long-tail distribution in spectral data due to the inconsistency of ore distribution and the scarcity of several natural minerals. To alleviate the interference of majority categories on minority categories, we propose Long-Tail Few-shot Module (LTFM) which is inspired by rethinking the fashionable decoupling strategy that conducts primary representation learning and further classifier retrained on mineral spectral data. In particular, LTFM serves as a multi-expert mode, where these experts are respectively specialized in improving feature representation learning, mitigating the long-tail effect, and alleviating the interference of few shots. Additionally, the loose coupling learning strategy is introduced to facilitate primary representation learning and the subsequent additional experts to inherit this knowledge. Experiments on two publicly available spectral datasets show that the proposed LTFM significantly outperforms existing methods. In the end, extensive ablation studies are conducted to investigate the effectiveness, correctness, and robustness of our proposal.

An efficient data fusion model based on Bayesian model averaging for robust water quality prediction using deep learning strategies

Accurate monitoring of dissolved oxygen (DO) levels is critical for stakeholders to effectively safeguard water resources and aquatic ecosystem health. This research presents an innovative data fusion framework based on Bayesian model averaging (BMA) by the combination of several neuroscience models (deep learning methodologies) including multilayer perceptron neural network (MLPNN), recurrent neural network (RNN), convolutional neural network (CNN), gated recurrent unit (GRU), long short-term memory (LSTM), and seasonal autoregressive integrated moving average with exogenous variables (SARIMAX). BMA has the capacity to greatly enhance the results attained by standalone approaches. In this study, two feature selection methods such as mutual information (MI) and recursive feature elimination (RFE) applied to select effective predictors and investigate the importance of each input parameters. The techniques were evaluated based on four different metrics and, finally, to demonstrate the usefulness and effectiveness of the newly implemented strategy, it was applied proficiently at the USGS stations, 01427510 and 02336152 within the USA. The findings from analyzing data based on two stations confirmed that the suggested approach (BMA) outperformed other methods such as MLPNN, RNN, CNN, LSTM, and SARIMAX when it came to predicting the levels of DO on a daily basis. In terms of RMSE, MAE and R2, BMA yielded 0.272 mg/L, 0.216 mg/L, and 0.975 using MI technique and 0.320 mg/L, 0.261 mg/L, and 0.965 using RFE method at 01427510 USGS station, respectively. Similarly, based on RMSE, MAE and R2, BMA produced DO prediction by RMSE = 0.352 mg/L, 0.264 mg/L, and 0.968 by MI approach and RMSE = 0.378 mg/L, 0.282 mg/L, and 0.963 via RFE process at 02336152 USGS station, respectively. After analyzing different combinations of input that obtained from feature selection paradigms, it was observed that the variables with the greatest impact on daily dissolved oxygen levels are the dissolved oxygen levels in the previous time period and the water temperature. On the other hand, the pH and turbidity have minimal influence on the daily DO. Finally, the results of this study confirmed that the BMA tool can be efficiently applied to predict DO concentration based on deep learning model outputs.

A Novel Feature Selection Method for Ultra High Dimensional Survival Data

Finding relevant features in ultra-high dimensional survival data is one of the most important and fundamental objectives in biology discovery and statistical acquisition. Conventional survival regression algorithms are challenged by the exponential increase in raw data. In real-world scenarios, data processing with ultra-high dimensionality has an impact, particularly on two-component structures like the kidneys, lungs, and eyes. Future system stability and the frequency of illness are both affected by gene interactions between two components. The traditional statistical procedures employed for the survival system are restricted to single component. To date, for ultra-high-dimensional survival data with two compartments, no feature selection method is available. Thus, with the goal to determine the optimal methods in this situation, this study suggested and contrasted the performance of ten variable selection approaches for ultra-high dimensional Renal Cell Carcinoma (RCC) survival data containing two compartments. The study attempted to combine Freund’s baseline hazard function as the baseline hazard of Cox model (Lasso Freund, Robust Lasso Freund, Elastic Net Freund) and integrated with sure independence screening (SIS) and iterative sure independence screening (ISIS) (i.e., LF-SIS, RLF-SIS, ENF-SIS, LF-ISIS, RLF-ISIS, ENF-ISIS) in an attempt to tackle this issue. Additionally, two basic approaches, LASSO and EN, were taken into consideration and EN is combined with SIS and ISIS (EN-SIS, EN-ISIS). Result shows that based on the validating model measures, including MSE (340.000), SSE (25300.0) and RMSE (16.490) suggest, the Robust Lasso Freund-Iterative Sure Independence Screening (RLF-ISIS) and Robust Lasso Freund-Sure Independence Screening (RLF-SIS) strategy performs superior to the other suggested approaches in terms of greater precision in picking variables. Though both methods showed lower R2 (0.71) which advocates the presence of the outliers in the dataset. Additionally, the box-plot of some selected predictive genes confirms the presence of outliers. Furthermore, two methods, RLF-ISIS and RLF-SIS, have been used to identify 49 and 68 genes that have both direct and indirect effects on patients with RCC. Finally, it can be concluded that although RLF-SIS and RLF-ISIS outperform other proposed approaches, they may, however, be regarded as a variable selection strategy but they might not be the optimal choice for ultra-high dimensional survival data with outliers. Nevertheless, the study can be expanded in the future by applying competitive risk theory to a sequential and parallel structure, which serves as the basis for most complex mechanical systems found in manufacturing facilities. Notably, no feature selection method is available for ultra-high-dimensional survival data with outliers and two-compartments. Therefore, to address this particular issue, further research should focus on developing an advanced hybrid feature selection approach, with a particular emphasis on deep learning strategies.

A novel classification framework for genome-wide association study of whole brain MRI images using deep learning.

Genome-wide association studies (GWASs) have been widely applied in the neuroimaging field to discover genetic variants associated with brain-related traits. So far, almost all GWASs conducted in neuroimaging genetics are performed on univariate quantitative features summarized from brain images. On the other hand, powerful deep learning technologies have dramatically improved our ability to classify images. In this study, we proposed and implemented a novel machine learning strategy for systematically identifying genetic variants that lead to detectable nuances on Magnetic Resonance Images (MRI). For a specific single nucleotide polymorphism (SNP), if MRI images labeled by genotypes of this SNP can be reliably distinguished using machine learning, we then hypothesized that this SNP is likely to be associated with brain anatomy or function which is manifested in MRI brain images. We applied this strategy to a catalog of MRI image and genotype data collected by the Alzheimer's Disease Neuroimaging Initiative (ADNI) consortium. From the results, we identified novel variants that show strong association to brain phenotypes.

The student-teacher experience as a model for students as partners (SaP) and for enhancing student engagement among Japanese students

Education in Japan and other Asian countries advocates the traditional passive style of learning where students learn through rote memorization in a teacher-centered environment. However, this is now changing due to globalization with Japan’s Ministry of Education promoting learning strategies to involve students in the learning process actively. Many studies involving students as partners have stated that students feel more at ease when they learn with and from their peers and that working in groups makes them learn significantly better and allows them to put into practice what they have taught to others or learned from others. This case study on the Student-Teacher Experience activity aims to present the concept of students as partners to enhance active learning in Japan and its possible application in other countries and fields of study with a similar learning environment as Japan. The activity allowed students to actively engage in class and enhanced learning.

Science of Learning Strategy Series: Article 6, Elaboration

Abstract: The science of learning (learning science) is an interprofessional field that concerns itself with how the brain learns and remembers important information. Learning science has compiled a set of evidence-based strategies, such as distributed practice, retrieval practice, and interleaving, which are quite relevant to continuing professional development. Spreading out study and practice separated by cognitive breaks (distributed practice), testing oneself to check mastery and memory of previously learned information (retrieval practice), and mixing the learning of separate but associated information (interleaving) represent strategies that are underutilized in continuing professional development. Participants and planners alike can benefit from learning science recommendations to inform their decisions. Elaboration, the subject of this article, is another evidence-based strategy that relates to how one studies or practices priority information.

THE USE OF ARTIFICIAL INTELLIGENCE, VIRTUAL AND AUGMENTED REALITY IN TEACHING GEOGRAPHY IN SECONDARY SCHOOL

In today's innovative education, artificial intelligence (AI), virtual reality (VR) and augmented reality (AR) technologies are increasingly emerging as powerful tools to enhance the learning experience, especially in the teaching of geography in primary schools. This research explores how these technologies are being used to improve geography teaching and stimulate students' interest. AI brings personalised learning where content is adapted to individual learners' needs. Easier access to information, better visualisation of content, and help with analysis and interpretation make learning more accessible for students. At the same time, the ease of recreating content and the help in tracking individual learners' progress allows teachers to adapt learning strategies better and provide additional resources according to individual needs. VR and AR technologies help to create more interactive and realistic learning environments. Virtual reality allows learners to experience geographical phenomena more directly, while augmented reality complements the real world with digital information that can enrich the understanding of geographical concepts. These technologies focus on creating interactive content, virtual field trips, simulations of natural phenomena, and encouraging student participation and interest. The study also discusses challenges and solutions related to introducing AI in school geography, such as access to technology, teacher training and professional development, curriculum adaptation and ethical dilemmas. AI, VR and AR technology enable a new, at least in Slovenia, technologically sophisticated approach to teaching geography in primary schools. It encourages students to explore, interact and better understand complex geographic content. Keywords: geography education, artificial intelligence, virtual reality, augmented reality, interactive content

WEKA: A Tool for Shaping Effective Student Learning Strategies for College School Leavers

As college school leavers transition into higher education or the workforce, they face numerous challenges in adapting to new learning environments and expectations. Effective learning strategies are crucial to ensuring their success in these new settings. This paper explores how WEKA, a powerful data mining tool, can be leveraged to shape and enhance student learning strategies for college school leavers. By analysing educational data, WEKA can provide insights into the most effective learning methods, helping educators tailor their approaches to meet the specific needs of these students. The paper discusses the application of WEKA in educational data mining, highlights its capabilities, and presents case studies demonstrating its effectiveness in shaping learning strategies.

"More than just a game, it's an app that builds awareness around Mental Health": Mental Health Stigma Reduction Using Games for Change

Although the significance of Mental Health (MH) is becoming increasingly more accepted worldwide, the level of MH Literacy (MHL) remains low. Individuals often find it uncomfortable to discuss this crucial topic. Unwillingness to discuss MH can be a result of the stigma surrounding the topic which results in misconceptions, discrimination, and reluctance to seek help or talk openly about MH challenges. Educating individuals about MH and encouraging open discussions are key to reducing stigma. Games have proven their effectiveness for diverse learning purposes. Our goal in this paper is to assess the potential of games and digital displays for MHL education and stigma reduction by incorporating MH educational content into a mobile game with different learning strategies and a large public display to create a community. We compare the pre and post-MHL of users before and after playing our game and the results show an improvement in the MHL of participants. Our paper contributes to game design approaches by identifying mechanisms for educating individuals about MH and our results indicate that the game helps reduce stigma through the use of large displays.

English Learning Strategies Supported by Wireless Communication Modules and Their Application in 5G

In this study, a simplified strategy for the English assistance model was provided to overcome the limitations of current school English display settings and increase the scope of 5G innovation. This study uses a learning strategy based convolutional neural network (LS-CNN) to determine the execution of college English learning strategies. This article breaks traditional school English learning strategies by integrating 5G, which significantly improves classroom communication. In addition, this development approach may improve the standards of English performance in various methods, while strictly and purposefully adhering to the display guidance of “displaying substitute actors based on their skills”. The early findings of the review revealed that the proposed method focuses on the weaknesses of the standard English display model used in universities, while improving the mental abilities of substitute actors in various fields, including listening, talking, reading, and composing.

Epsilon-Greedy Thompson Sampling to Bayesian Optimization

Abstract Bayesian optimization (BO) has become a powerful tool for solving simulation-based engineering optimization problems thanks to its ability to integrate physical and mathematical understandings, consider uncertainty, and address the exploitation–exploration dilemma. Thompson sampling (TS) is a preferred solution for BO to handle the exploitation–exploration trade-off. While it prioritizes exploration by generating and minimizing random sample paths from probabilistic models—a fundamental ingredient of BO—TS weakly manages exploitation by gathering information about the true objective function after it obtains new observations. In this work, we improve the exploitation of TS by incorporating the e-greedy policy, a well-established selection strategy in reinforcement learning. We first delineate two extremes of TS, namely the generic TS and the sample-average TS. The former promotes exploration, while the latter favors exploitation. We then adopt the e-greedy policy to randomly switch between these two extremes. Small and large values of e govern exploitation and exploration, respectively. By minimizing two benchmark functions and solving an inverse problem of a steel cantilever beam, we empirically show that e-greedy TS equipped with an appropriate e is more robust than its two extremes, matching or outperforming the better of the generic TS and the sample-average TS.

Group work and translanguaging as coping strategies for content learning in English-medium higher education contexts

Abstract English as a medium of instruction (EMI) has been increasingly used in Higher Education institutions in countries where English is spoken as a second or foreign language (ESL/EFL). Research over the last decade has predominantly focused on EMI implementation, perceptions, and attitudes of stakeholders towards EMI as well as the challenges experienced by both teachers and students. However, little research, thus far, has focused on the coping strategies used by students to overcome challenges and manage their EMI study. The present study duly aims to fill this gap and contribute to the advancement of EMI research and pedagogy in this under-researched area. The study particularly attempts to explore group work and translanguaging as coping strategies that students employ to overcome content learning challenges and succeed in their EMI study. The study adopted a qualitative methodology and semi-structured interviews were used as the main method of data collection. Twenty participants voluntarily took part in the study. The data were analysed thematically and inductively. Results showed that students utilised several coping strategies to overcome EMI learning challenges and described these strategies as being helpful for them to understand EMI subject content. The pedagogical implications are discussed.