Student-centered learning approach is focused on the students' demands and interests.Applying student-centered approach puts forward higher requirement to English teachers.This article first analyzes the theory of...Student-centered learning approach is focused on the students' demands and interests.Applying student-centered approach puts forward higher requirement to English teachers.This article first analyzes the theory of student-centered learning approach and compares teacher-centered approach with it.Based on the research information and teaching experience,the author summarizes four strategies about how to apply student-centered learning approach to English listening and speaking class in vocational schools.展开更多
The exploration of high-performance materials presents a fundamental challenge in materials science,particularly in predicting properties for materials beyond the known range of target property values(extrapolation).T...The exploration of high-performance materials presents a fundamental challenge in materials science,particularly in predicting properties for materials beyond the known range of target property values(extrapolation).This study formally investigated the interpolation-extrapolation trade-off phenomenon in the prediction capabilities of machine learning(ML)models.A new ML scheme was proposed,featuring a newly developed ML model and forward cross-validation-based hyperparameter optimization,which demonstrated superior extrapolation prediction across multiple materials datasets.Based on this ML scheme,multi-objective optimization was performed to systematically identify lightweight Mg-Zn-Al alloys with both high bulk modulus and high Debye temperature.Subsequently,the designed alloys were validated through density functional theory calculations.Furthermore,a three-category classification strategy was summarized through the dual-driven approach combining domain knowledge and data,emphasizing their synergistic potential for materials discovery.The practical framework developed in this study provides a novel research perspective for exploring high-performance materials.展开更多
Deep learning algorithms have been rapidly incorporated into many different applications due to the increase in computational power and the availability of massive amounts of data.Recently,both deep learning and ensem...Deep learning algorithms have been rapidly incorporated into many different applications due to the increase in computational power and the availability of massive amounts of data.Recently,both deep learning and ensemble learning have been used to recognize underlying structures and patterns from high-level features to make predictions/decisions.With the growth in popularity of deep learning and ensemble learning algorithms,they have received significant attention from both scientists and the industrial community due to their superior ability to learn features from big data.Ensemble deep learning has exhibited significant performance in enhancing learning generalization through the use of multiple deep learning algorithms.Although ensemble deep learning has large quantities of training parameters,which results in time and space overheads,it performs much better than traditional ensemble learning.Ensemble deep learning has been successfully used in several areas,such as bioinformatics,finance,and health care.In this paper,we review and investigate recent ensemble deep learning algorithms and techniques in health care domains,medical imaging,health care data analytics,genomics,diagnosis,disease prevention,and drug discovery.We cover several widely used deep learning algorithms along with their architectures,including deep neural networks(DNNs),convolutional neural networks(CNNs),recurrent neural networks(RNNs),and generative adversarial networks(GANs).Common healthcare tasks,such as medical imaging,electronic health records,and genomics,are also demonstrated.Furthermore,in this review,the challenges inherent in reducing the burden on the healthcare system are discussed and explored.Finally,future directions and opportunities for enhancing healthcare model performance are discussed.展开更多
Online interactive learning plays a crucial role in improving online education quality.This grounded theory study examines:(1)what key factors shape EFL learners’online interactive learning,(2)how these factors form ...Online interactive learning plays a crucial role in improving online education quality.This grounded theory study examines:(1)what key factors shape EFL learners’online interactive learning,(2)how these factors form an empirically validated model,and(3)how they interact within this model,through systematic analysis of 9,207 discussion forum posts from a Chinese University MOOC platform.Results demonstrate that learning drive,course structure,teaching competence,interaction behavior,expected outcomes,and online learning context significantly influence EFL online interactive learning.The analysis reveals two key mechanisms:expected outcomes mediate the effects of learning drive(β=0.45),course structure,teaching competence,and interaction behavior(β=0.35)on learning outcomes,while online learning context moderates these relationships(β=0.25).Specifically,learning drive provides intrinsic/extrinsic motivation,whereas course structure,teaching competence,interaction behavior,and expected outcomes collectively enhance interaction quality and sustainability.These findings,derived through rigorous grounded theory methodology involving open,axial,and selective coding of large-scale interaction data,yield three key contributions:(1)a comprehensive theoretical model of EFL online learning dynamics,(2)empirical validation of mediation/moderation mechanisms,and(3)practical strategies for designing scaffolded interaction protocols and adaptive feedback systems.The study establishes that its theoretically saturated model(achieved after analyzing 7,366 posts with 1,841 verification cases)offers educators evidence-based approaches to optimize collaborative interaction in digital EFL environments.展开更多
Contemporary higher education prioritizes cultivating students’key competencies and comprehensive problem-solving abilities,specifically fostering innovation,goal orientation,and initiative.This study investigates a ...Contemporary higher education prioritizes cultivating students’key competencies and comprehensive problem-solving abilities,specifically fostering innovation,goal orientation,and initiative.This study investigates a pedagogical framework that synergizes Research-Led Learning(RLL)and Project-Based Learning(PBL)to establish an open,exploratory learning environment.Employing a case study methodology,the research tracked architecture students engaging in a structured PBL process involving rigorous research activities—ranging from theoretical analysis to field investigations—to develop evidence-based design solutions.Evaluations from both student and faculty perspectives assessed the pedagogical effectiveness regarding learning outcomes and competency development.The findings indicate that this methodology effectively bridges the gap between research and practice,significantly bolstering students’capacity to address authentic challenges and propelling self-directed learning in architectural education.展开更多
Enhancing the mechanical properties is crucial for polyimide films,but the mechanical properties(Young's modulus,tensile strength,and elongation at break)mutually constrain each other,complicating simultaneous enh...Enhancing the mechanical properties is crucial for polyimide films,but the mechanical properties(Young's modulus,tensile strength,and elongation at break)mutually constrain each other,complicating simultaneous enhancement via traditional trial-and-error methods.In this work,we proposed a materials genome approach to design and screen phenylethynyl-terminated polyimides for films with enhanced mechani-cal properties.We first established machine learning models to predict Young's modulus,tensile strength,and elongation at break to explore the chemical space containing thousands of candidate structures.The accuracies of the machine learning models were verified by molecular dynamics simulations on screened polyimides and experimental testing on three representative polyimide films.The performance advantages of the best-selected polyimides were analyzed by comparing well-known polyimides based on molecular dynamics simulations,and the structural rationale was revealed by"gene"analysis and feature importance evaluation.This work provides a cost-effective strategy for designing polyimide films withenhancedmechanical properties.展开更多
In response to the misconception that Communicative Language Teaching means no teaching of grammar,it is argued that grammar is as important as traffic rules for safe and smooth traffic on the road.To achieve appropri...In response to the misconception that Communicative Language Teaching means no teaching of grammar,it is argued that grammar is as important as traffic rules for safe and smooth traffic on the road.To achieve appropriate and effective communication,a communicative approach to college grammar teaching and learning is proposed.Both teachers and learners should change their attitudes toward and conceptions about grammar teaching and learning;additionally,teaching grammar in the company of reading and writing helps learners learn and acquire grammar in meaningful contexts.展开更多
Objectives： This study aimed to compare the learning curves of percutaneous endoscopic lumbar discectomy （PELD） in a transforaminal approach at the L4/5 and L5/S1 levels. Methods： We retrospectively reviewed the f...Objectives： This study aimed to compare the learning curves of percutaneous endoscopic lumbar discectomy （PELD） in a transforaminal approach at the L4/5 and L5/S1 levels. Methods： We retrospectively reviewed the first 60 cases at the L4/5 level （Group I） and the first 60 cases at the L5/S1 level （Group II） of PELD performed by one spine surgeon. The patients were divided into subgroups A, B, and C （Group I： A cases 1-20, B cases 21-40, C cases 41-60; Group I1： A cases 1-20, B cases 21-40, C cases 41-60）. Operation time was thoroughly analyzed. Results： Compared with the L4/5 level, the learning curve of transforaminal PELD at the L5/S1 level was flatter. The mean operation times of Groups IA, IB, and IC were （88.75±17.02）, （67.75±6.16）, and （64.85±7.82） min, respectively. There was a significant difference between Groups A and B （P〈0.05）, but no significant difference between Groups B and C （P=-0.20）. The mean operation times of Groups IIA, liB, and IIC were （117.25±13.62）, （109.50±11.20）, and （92.15±11.94） rain, respectively. There was no significant difference between Groups A and B （P=0.06）, but there was a significant difference between Groups B and C （P〈0.05）. There were 6 cases of postoperative dysesthesia （POD） in Group I and 2 cases in Group IIA （P=-0.27）. There were 2 cases of residual disc in Group I, and 4 cases in Group II （P=0.67）. There were 3 cases of recurrence in Group I, and 2 cases in Group II （P〉0.05）. Conclusions： Compared with the L5/S1 level, the learning curve of PELD in a transforaminal approach at the L4/5 level was steeper, suggesting that the L4/5 level might be easier to master after short-term professional training.展开更多
This paper deals with the iterative learning control （ILC） design for multiple-input multiple-output （MIMO）,time-delay systems （TDS）.Two feedback ILC schemes are considered using the so-called two-dimensional ...This paper deals with the iterative learning control （ILC） design for multiple-input multiple-output （MIMO）,time-delay systems （TDS）.Two feedback ILC schemes are considered using the so-called two-dimensional （2D） analysis approach.It shows that continuous-discrete 2D Roesser systems can be developed to describe the entire learning dynamics of both ILC schemes,based on which necessary and sufficient conditions for their stability can be provided.A numerical example is included to validate the theoretical analysis.展开更多
In the rapidly evolving landscape of television advertising,optimizing ad schedules to maximize viewer engagement and revenue has become significant.Traditional methods often operate in silos,limiting the potential in...In the rapidly evolving landscape of television advertising,optimizing ad schedules to maximize viewer engagement and revenue has become significant.Traditional methods often operate in silos,limiting the potential insights gained from broader data analysis due to concerns over privacy and data sharing.This article introduces a novel approach that leverages Federated Learning(FL)to enhance TV ad schedule optimization,combining the strengths of local optimization techniques with the power of global Machine Learning(ML)models to uncover actionable insights without compromising data privacy.It combines linear programming for initial ads scheduling optimization with ML—specifically,a K-Nearest Neighbors(KNN)model—for predicting ad spot positions.Taking into account the diversity and the difficulty of the ad-scheduling problem,we propose a prescriptivepredictive approach in which first the position of the ads is optimized(using Google’s OR-Tools CP-SAT)and then the scheduled position of all ads will be the result of the optimization problem.Second,this output becomes the target of a predictive task that predicts the position of new entries based on their characteristics ensuring the implementation of the scheduling at large scale(using KNN,Light Gradient Boosting Machine and Random Forest).Furthermore,we explore the integration of FL to enhance predictive accuracy and strategic insight across different broadcasting networks while preserving data privacy.The FL approach resulted in 8750 ads being precisely matched to their optimal category placements,showcasing an alignment with the intended diversity objectives.Additionally,there was a minimal deviation observed,with 1133 ads positioned within a one-category variance from their ideal placement in the original dataset.展开更多
In today's modern electric vehicles,enhancing the safety-critical cyber-physical system(CPS)'s performance is necessary for the safe maneuverability of the vehicle.As a typical CPS,the braking system is crucia...In today's modern electric vehicles,enhancing the safety-critical cyber-physical system(CPS)'s performance is necessary for the safe maneuverability of the vehicle.As a typical CPS,the braking system is crucial for the vehicle design and safe control.However,precise state estimation of the brake pressure is desired to perform safe driving with a high degree of autonomy.In this paper,a sensorless state estimation technique of the vehicle's brake pressure is developed using a deep-learning approach.A deep neural network(DNN)is structured and trained using deep-learning training techniques,such as,dropout and rectified units.These techniques are utilized to obtain more accurate model for brake pressure state estimation applications.The proposed model is trained using real experimental training data which were collected via conducting real vehicle testing.The vehicle was attached to a chassis dynamometer while the brake pressure data were collected under random driving cycles.Based on these experimental data,the DNN is trained and the performance of the proposed state estimation approach is validated accordingly.The results demonstrate high-accuracy brake pressure state estimation with RMSE of 0.048 MPa.展开更多
We consider an image semantic communication system in a time-varying fading Gaussian MIMO channel,with a finite number of channel states.A deep learning-aided broadcast approach scheme is proposed to benefit the adapt...We consider an image semantic communication system in a time-varying fading Gaussian MIMO channel,with a finite number of channel states.A deep learning-aided broadcast approach scheme is proposed to benefit the adaptive semantic transmission in terms of different channel states.We combine the classic broadcast approach with the image transformer to implement this adaptive joint source and channel coding(JSCC)scheme.Specifically,we utilize the neural network(NN)to jointly optimize the hierarchical image compression and superposition code mapping within this scheme.The learned transformers and codebooks allow recovering of the image with an adaptive quality and low error rate at the receiver side,in each channel state.The simulation results exhibit our proposed scheme can dynamically adapt the coding to the current channel state and outperform some existing intelligent schemes with the fixed coding block.展开更多
L1 and L2 acquisition, in some respects, are similar. Language development in children goes hand in hand with physical and cognitive development. Children learn their first language by imitation, but not always and no...L1 and L2 acquisition, in some respects, are similar. Language development in children goes hand in hand with physical and cognitive development. Children learn their first language by imitation, but not always and not only by imitation. There seems to be some "innate capacities" that make children start to speak at the same time they do and in the way they do it. Adults learning a second language usually are controlled more by their motivation. But language input is important for both L1 and L2 acquisition. Though there are differences between CL1 and between CL2 and AL2, the way in which these learners acquire some of the grammatical morphemes is similar. This, together with some other evidence, shows that it is not only children who can acquire language. Adults can also acquire a language. But when adults acquire a language, they should also learn it. Some of the ways in which children acquire their language can be used as a model for L2 acquisition, even for Chinese students whose language is unrelated to English and whose culture is different. Learning the culture of the English-speaking countries will benefit the learning of the language. Like children, listening should also be well in advance of speaking in L2 acquisition. To train listening comprehension skills, Asher’s TPR approach proves more effective. TPR approach is at the moment limited to the beginning stage only. In order for students to gain all the five skills in a second language learning, namely, listening, speaking, reading, writing, and interpreting/translating, other methods should be used at the same time, or at later stages.展开更多
DEAR EDITOR,Somatic mutations are a large category of genetic variations,which play an essential role in tumorigenesis. Detection of somatic single nucleotide variants(SNVs) could facilitate downstream analysis of tum...DEAR EDITOR,Somatic mutations are a large category of genetic variations,which play an essential role in tumorigenesis. Detection of somatic single nucleotide variants(SNVs) could facilitate downstream analysis of tumorigenesis. Many computational methods have been developed to detect SNVs, but most require normal matched samples to differentiate somatic SNVs from the normal state, which can be difficult to obtain.展开更多
The Approaches to Learning addresses how children learn-this includes children’s attitudes and interests in learning.This domain reflects behaviours and attitudes such as curiosity,problem-solving,maintaining attenti...The Approaches to Learning addresses how children learn-this includes children’s attitudes and interests in learning.This domain reflects behaviours and attitudes such as curiosity,problem-solving,maintaining attention and persistence.The research study focused on examining the fathers’parenting practices and the children’s approaches to learning from three through five years.The study used a cross sectional research design and data was generated using focal group discussions,interview guides and child behaviour rating scale on how fathers’parenting practices contribute to children’s approaches to learning.Results revealed that,Fathers’parenting practices and Children’s curiosity were found to have a very positive relationship(r=0.396,p<0.05).Fathers’parenting practices and children’s learning were found to have a significant positive relationship(r=0.420,p<0.05).Findings also indicated that fathers’parenting practices and children’s creativity were found to have an average positive relationship(r=0.379,p<0.05).Arising out of the findings,the study recommended that fathers’parenting programs be put in place to help them up bring the child in holistic manner.展开更多
Cardiovascular disease is among the top five fatal diseases that affect lives worldwide.Therefore,its early prediction and detection are crucial,allowing one to take proper and necessary measures at earlier stages.Mac...Cardiovascular disease is among the top five fatal diseases that affect lives worldwide.Therefore,its early prediction and detection are crucial,allowing one to take proper and necessary measures at earlier stages.Machine learning(ML)techniques are used to assist healthcare providers in better diagnosing heart disease.This study employed three boosting algorithms,namely,gradient boost,XGBoost,and AdaBoost,to predict heart disease.The dataset contained heart disease-related clinical features and was sourced from the publicly available UCI ML repository.Exploratory data analysis is performed to find the characteristics of data samples about descriptive and inferential statistics.Specifically,it was carried out to identify and replace outliers using the interquartile range and detect and replace the missing values using the imputation method.Results were recorded before and after the data preprocessing techniques were applied.Out of all the algorithms,gradient boosting achieved the highest accuracy rate of 92.20%for the proposed model.The proposed model yielded better results with gradient boosting in terms of precision,recall,and f1-score.It attained better prediction performance than the existing works and can be used for other diseases that share common features using transfer learning.展开更多
Electromagnetic sandwich metastructure(ESM)consisting of different functional layers,has gained in-creasing attention in radiation prevention and radar stealth.However,the current ESM design is primar-ily based on the...Electromagnetic sandwich metastructure(ESM)consisting of different functional layers,has gained in-creasing attention in radiation prevention and radar stealth.However,the current ESM design is primar-ily based on the separation design method,ignoring electromagnetic-mechanical interactions between layers.Thus,subject to thin thickness constraint of ESM,it is a great challenge to achieve broadband microwave absorption(MA)and excellent mechanical performance simultaneously.To address this is-sue,an electromagnetic-mechanical collaborative design approach was proposed for ESM.The relations of geometric-electromagnetic and geometric-mechanical of ESM were first identified by machine learn-ing.They were then integrated with the heuristic genetic optimization algorithm to perform the highly efficient design.The designed ESM can achieve 36.4 GHz effective absorption bandwidth(EAB,RL≤-10 dB),334.3 MPa equivalent bending strength and 83 MPa compressive strength with a thickness of 9.3 mm,possessing the widest EAB and highest bending strength within the current available MA struc-tures(thickness less than 9.5 mm).The proposed approach provides an efficient tool for the design of electromagnetic-mechanical optimal ESM.展开更多
Low-voltage direct current(DC)microgrids have recently emerged as a promising and viable alternative to traditional alternating cur-rent(AC)microgrids,offering numerous advantages.Consequently,researchers are explorin...Low-voltage direct current(DC)microgrids have recently emerged as a promising and viable alternative to traditional alternating cur-rent(AC)microgrids,offering numerous advantages.Consequently,researchers are exploring the potential of DC microgrids across var-ious configurations.However,despite the sustainability and accuracy offered by DC microgrids,they pose various challenges when integrated into modern power distribution systems.Among these challenges,fault diagnosis holds significant importance.Rapid fault detection in DC microgrids is essential to maintain stability and ensure an uninterrupted power supply to critical loads.A primary chal-lenge is the lack of standards and guidelines for the protection and safety of DC microgrids,including fault detection,location,and clear-ing procedures for both grid-connected and islanded modes.In response,this study presents a brief overview of various approaches for protecting DC microgrids.展开更多
The prediction of pregnancy-related hazards must be accurate and timely to safeguard mother and fetal health.This study aims to enhance risk prediction in pregnancywith a novel deep learningmodel based on a Long Short...The prediction of pregnancy-related hazards must be accurate and timely to safeguard mother and fetal health.This study aims to enhance risk prediction in pregnancywith a novel deep learningmodel based on a Long Short-Term Memory(LSTM)generator,designed to capture temporal relationships in cardiotocography(CTG)data.This methodology integrates CTG signals with demographic characteristics and utilizes preprocessing techniques such as noise reduction,normalization,and segmentation to create high-quality input for themodel.It uses convolutional layers to extract spatial information,followed by LSTM layers to model sequences for superior predictive performance.The overall results show that themodel is robust,with an accuracy of 91.5%,precision of 89.8%,recall of 90.4%,and F1-score of 90.1%that outperformed the corresponding baselinemodels,CNN(Convolutional Neural Network)and traditional RNN(Recurrent Neural Network),by 2.3%and 6.1%,respectively.Rather,the ability to detect pregnancy-related abnormalities has considerable therapeutic potential,with the possibility for focused treatments and individualized maternal healthcare approaches,the research team concluded.展开更多
文摘Student-centered learning approach is focused on the students' demands and interests.Applying student-centered approach puts forward higher requirement to English teachers.This article first analyzes the theory of student-centered learning approach and compares teacher-centered approach with it.Based on the research information and teaching experience,the author summarizes four strategies about how to apply student-centered learning approach to English listening and speaking class in vocational schools.
基金supported by National Natural Science Foundation of China(No.51671075 and 51971086)Natural Science Foundation of Heilongjiang Province of China(No.LH2022E081).
文摘The exploration of high-performance materials presents a fundamental challenge in materials science,particularly in predicting properties for materials beyond the known range of target property values(extrapolation).This study formally investigated the interpolation-extrapolation trade-off phenomenon in the prediction capabilities of machine learning(ML)models.A new ML scheme was proposed,featuring a newly developed ML model and forward cross-validation-based hyperparameter optimization,which demonstrated superior extrapolation prediction across multiple materials datasets.Based on this ML scheme,multi-objective optimization was performed to systematically identify lightweight Mg-Zn-Al alloys with both high bulk modulus and high Debye temperature.Subsequently,the designed alloys were validated through density functional theory calculations.Furthermore,a three-category classification strategy was summarized through the dual-driven approach combining domain knowledge and data,emphasizing their synergistic potential for materials discovery.The practical framework developed in this study provides a novel research perspective for exploring high-performance materials.
基金funded by Taif University,Saudi Arabia,project No.(TU-DSPP-2024-263).
文摘Deep learning algorithms have been rapidly incorporated into many different applications due to the increase in computational power and the availability of massive amounts of data.Recently,both deep learning and ensemble learning have been used to recognize underlying structures and patterns from high-level features to make predictions/decisions.With the growth in popularity of deep learning and ensemble learning algorithms,they have received significant attention from both scientists and the industrial community due to their superior ability to learn features from big data.Ensemble deep learning has exhibited significant performance in enhancing learning generalization through the use of multiple deep learning algorithms.Although ensemble deep learning has large quantities of training parameters,which results in time and space overheads,it performs much better than traditional ensemble learning.Ensemble deep learning has been successfully used in several areas,such as bioinformatics,finance,and health care.In this paper,we review and investigate recent ensemble deep learning algorithms and techniques in health care domains,medical imaging,health care data analytics,genomics,diagnosis,disease prevention,and drug discovery.We cover several widely used deep learning algorithms along with their architectures,including deep neural networks(DNNs),convolutional neural networks(CNNs),recurrent neural networks(RNNs),and generative adversarial networks(GANs).Common healthcare tasks,such as medical imaging,electronic health records,and genomics,are also demonstrated.Furthermore,in this review,the challenges inherent in reducing the burden on the healthcare system are discussed and explored.Finally,future directions and opportunities for enhancing healthcare model performance are discussed.
文摘Online interactive learning plays a crucial role in improving online education quality.This grounded theory study examines:(1)what key factors shape EFL learners’online interactive learning,(2)how these factors form an empirically validated model,and(3)how they interact within this model,through systematic analysis of 9,207 discussion forum posts from a Chinese University MOOC platform.Results demonstrate that learning drive,course structure,teaching competence,interaction behavior,expected outcomes,and online learning context significantly influence EFL online interactive learning.The analysis reveals two key mechanisms:expected outcomes mediate the effects of learning drive(β=0.45),course structure,teaching competence,and interaction behavior(β=0.35)on learning outcomes,while online learning context moderates these relationships(β=0.25).Specifically,learning drive provides intrinsic/extrinsic motivation,whereas course structure,teaching competence,interaction behavior,and expected outcomes collectively enhance interaction quality and sustainability.These findings,derived through rigorous grounded theory methodology involving open,axial,and selective coding of large-scale interaction data,yield three key contributions:(1)a comprehensive theoretical model of EFL online learning dynamics,(2)empirical validation of mediation/moderation mechanisms,and(3)practical strategies for designing scaffolded interaction protocols and adaptive feedback systems.The study establishes that its theoretically saturated model(achieved after analyzing 7,366 posts with 1,841 verification cases)offers educators evidence-based approaches to optimize collaborative interaction in digital EFL environments.
基金received approval from a committee named Innovation Institute for Sustainable Maritime Architecture Research and Technology(iSMART)(The certificate number was 2022-5-22-01).
文摘Contemporary higher education prioritizes cultivating students’key competencies and comprehensive problem-solving abilities,specifically fostering innovation,goal orientation,and initiative.This study investigates a pedagogical framework that synergizes Research-Led Learning(RLL)and Project-Based Learning(PBL)to establish an open,exploratory learning environment.Employing a case study methodology,the research tracked architecture students engaging in a structured PBL process involving rigorous research activities—ranging from theoretical analysis to field investigations—to develop evidence-based design solutions.Evaluations from both student and faculty perspectives assessed the pedagogical effectiveness regarding learning outcomes and competency development.The findings indicate that this methodology effectively bridges the gap between research and practice,significantly bolstering students’capacity to address authentic challenges and propelling self-directed learning in architectural education.
基金supported by the National Key R&D Program of China(No.2022YFB3707302)the National Natural Science Foundation of China(Nos.52394271 , 52394270).
文摘Enhancing the mechanical properties is crucial for polyimide films,but the mechanical properties(Young's modulus,tensile strength,and elongation at break)mutually constrain each other,complicating simultaneous enhancement via traditional trial-and-error methods.In this work,we proposed a materials genome approach to design and screen phenylethynyl-terminated polyimides for films with enhanced mechani-cal properties.We first established machine learning models to predict Young's modulus,tensile strength,and elongation at break to explore the chemical space containing thousands of candidate structures.The accuracies of the machine learning models were verified by molecular dynamics simulations on screened polyimides and experimental testing on three representative polyimide films.The performance advantages of the best-selected polyimides were analyzed by comparing well-known polyimides based on molecular dynamics simulations,and the structural rationale was revealed by"gene"analysis and feature importance evaluation.This work provides a cost-effective strategy for designing polyimide films withenhancedmechanical properties.
文摘In response to the misconception that Communicative Language Teaching means no teaching of grammar,it is argued that grammar is as important as traffic rules for safe and smooth traffic on the road.To achieve appropriate and effective communication,a communicative approach to college grammar teaching and learning is proposed.Both teachers and learners should change their attitudes toward and conceptions about grammar teaching and learning;additionally,teaching grammar in the company of reading and writing helps learners learn and acquire grammar in meaningful contexts.
文摘Objectives： This study aimed to compare the learning curves of percutaneous endoscopic lumbar discectomy （PELD） in a transforaminal approach at the L4/5 and L5/S1 levels. Methods： We retrospectively reviewed the first 60 cases at the L4/5 level （Group I） and the first 60 cases at the L5/S1 level （Group II） of PELD performed by one spine surgeon. The patients were divided into subgroups A, B, and C （Group I： A cases 1-20, B cases 21-40, C cases 41-60; Group I1： A cases 1-20, B cases 21-40, C cases 41-60）. Operation time was thoroughly analyzed. Results： Compared with the L4/5 level, the learning curve of transforaminal PELD at the L5/S1 level was flatter. The mean operation times of Groups IA, IB, and IC were （88.75±17.02）, （67.75±6.16）, and （64.85±7.82） min, respectively. There was a significant difference between Groups A and B （P〈0.05）, but no significant difference between Groups B and C （P=-0.20）. The mean operation times of Groups IIA, liB, and IIC were （117.25±13.62）, （109.50±11.20）, and （92.15±11.94） rain, respectively. There was no significant difference between Groups A and B （P=0.06）, but there was a significant difference between Groups B and C （P〈0.05）. There were 6 cases of postoperative dysesthesia （POD） in Group I and 2 cases in Group IIA （P=-0.27）. There were 2 cases of residual disc in Group I, and 4 cases in Group II （P=0.67）. There were 3 cases of recurrence in Group I, and 2 cases in Group II （P〉0.05）. Conclusions： Compared with the L5/S1 level, the learning curve of PELD in a transforaminal approach at the L4/5 level was steeper, suggesting that the L4/5 level might be easier to master after short-term professional training.
基金supported by the National Natural Science Foundation of China(No.60727002,60774003,60921001,90916024)the COSTIND(No.A2120061303)the National 973 Program(No.2005CB321902)
文摘This paper deals with the iterative learning control （ILC） design for multiple-input multiple-output （MIMO）,time-delay systems （TDS）.Two feedback ILC schemes are considered using the so-called two-dimensional （2D） analysis approach.It shows that continuous-discrete 2D Roesser systems can be developed to describe the entire learning dynamics of both ILC schemes,based on which necessary and sufficient conditions for their stability can be provided.A numerical example is included to validate the theoretical analysis.
基金supported by a grant of the Ministry of Research,Innovation and Digitization,CNCS/CCCDI-UEFISCDI,project number COFUND-DUT-OPEN4CEC-1,within PNCDI IV.
文摘In the rapidly evolving landscape of television advertising,optimizing ad schedules to maximize viewer engagement and revenue has become significant.Traditional methods often operate in silos,limiting the potential insights gained from broader data analysis due to concerns over privacy and data sharing.This article introduces a novel approach that leverages Federated Learning(FL)to enhance TV ad schedule optimization,combining the strengths of local optimization techniques with the power of global Machine Learning(ML)models to uncover actionable insights without compromising data privacy.It combines linear programming for initial ads scheduling optimization with ML—specifically,a K-Nearest Neighbors(KNN)model—for predicting ad spot positions.Taking into account the diversity and the difficulty of the ad-scheduling problem,we propose a prescriptivepredictive approach in which first the position of the ads is optimized(using Google’s OR-Tools CP-SAT)and then the scheduled position of all ads will be the result of the optimization problem.Second,this output becomes the target of a predictive task that predicts the position of new entries based on their characteristics ensuring the implementation of the scheduling at large scale(using KNN,Light Gradient Boosting Machine and Random Forest).Furthermore,we explore the integration of FL to enhance predictive accuracy and strategic insight across different broadcasting networks while preserving data privacy.The FL approach resulted in 8750 ads being precisely matched to their optimal category placements,showcasing an alignment with the intended diversity objectives.Additionally,there was a minimal deviation observed,with 1133 ads positioned within a one-category variance from their ideal placement in the original dataset.
文摘In today's modern electric vehicles,enhancing the safety-critical cyber-physical system(CPS)'s performance is necessary for the safe maneuverability of the vehicle.As a typical CPS,the braking system is crucial for the vehicle design and safe control.However,precise state estimation of the brake pressure is desired to perform safe driving with a high degree of autonomy.In this paper,a sensorless state estimation technique of the vehicle's brake pressure is developed using a deep-learning approach.A deep neural network(DNN)is structured and trained using deep-learning training techniques,such as,dropout and rectified units.These techniques are utilized to obtain more accurate model for brake pressure state estimation applications.The proposed model is trained using real experimental training data which were collected via conducting real vehicle testing.The vehicle was attached to a chassis dynamometer while the brake pressure data were collected under random driving cycles.Based on these experimental data,the DNN is trained and the performance of the proposed state estimation approach is validated accordingly.The results demonstrate high-accuracy brake pressure state estimation with RMSE of 0.048 MPa.
基金supported in part by the National Key R&D Project of China under Grant 2020YFA0712300National Natural Science Foundation of China under Grant NSFC-62231022,12031011supported in part by the NSF of China under Grant 62125108。
文摘We consider an image semantic communication system in a time-varying fading Gaussian MIMO channel,with a finite number of channel states.A deep learning-aided broadcast approach scheme is proposed to benefit the adaptive semantic transmission in terms of different channel states.We combine the classic broadcast approach with the image transformer to implement this adaptive joint source and channel coding(JSCC)scheme.Specifically,we utilize the neural network(NN)to jointly optimize the hierarchical image compression and superposition code mapping within this scheme.The learned transformers and codebooks allow recovering of the image with an adaptive quality and low error rate at the receiver side,in each channel state.The simulation results exhibit our proposed scheme can dynamically adapt the coding to the current channel state and outperform some existing intelligent schemes with the fixed coding block.
文摘L1 and L2 acquisition, in some respects, are similar. Language development in children goes hand in hand with physical and cognitive development. Children learn their first language by imitation, but not always and not only by imitation. There seems to be some "innate capacities" that make children start to speak at the same time they do and in the way they do it. Adults learning a second language usually are controlled more by their motivation. But language input is important for both L1 and L2 acquisition. Though there are differences between CL1 and between CL2 and AL2, the way in which these learners acquire some of the grammatical morphemes is similar. This, together with some other evidence, shows that it is not only children who can acquire language. Adults can also acquire a language. But when adults acquire a language, they should also learn it. Some of the ways in which children acquire their language can be used as a model for L2 acquisition, even for Chinese students whose language is unrelated to English and whose culture is different. Learning the culture of the English-speaking countries will benefit the learning of the language. Like children, listening should also be well in advance of speaking in L2 acquisition. To train listening comprehension skills, Asher’s TPR approach proves more effective. TPR approach is at the moment limited to the beginning stage only. In order for students to gain all the five skills in a second language learning, namely, listening, speaking, reading, writing, and interpreting/translating, other methods should be used at the same time, or at later stages.
基金supported by the CAS Pioneer Hundred Talents Program and National Natural Science Foundation of China (32070683) to Y.P.C。
文摘DEAR EDITOR,Somatic mutations are a large category of genetic variations,which play an essential role in tumorigenesis. Detection of somatic single nucleotide variants(SNVs) could facilitate downstream analysis of tumorigenesis. Many computational methods have been developed to detect SNVs, but most require normal matched samples to differentiate somatic SNVs from the normal state, which can be difficult to obtain.
文摘The Approaches to Learning addresses how children learn-this includes children’s attitudes and interests in learning.This domain reflects behaviours and attitudes such as curiosity,problem-solving,maintaining attention and persistence.The research study focused on examining the fathers’parenting practices and the children’s approaches to learning from three through five years.The study used a cross sectional research design and data was generated using focal group discussions,interview guides and child behaviour rating scale on how fathers’parenting practices contribute to children’s approaches to learning.Results revealed that,Fathers’parenting practices and Children’s curiosity were found to have a very positive relationship(r=0.396,p<0.05).Fathers’parenting practices and children’s learning were found to have a significant positive relationship(r=0.420,p<0.05).Findings also indicated that fathers’parenting practices and children’s creativity were found to have an average positive relationship(r=0.379,p<0.05).Arising out of the findings,the study recommended that fathers’parenting programs be put in place to help them up bring the child in holistic manner.
基金This work was supported by National Research Foundation of Korea-Grant funded by the Korean Government(MSIT)-NRF-2020R1A2B5B02002478.
文摘Cardiovascular disease is among the top five fatal diseases that affect lives worldwide.Therefore,its early prediction and detection are crucial,allowing one to take proper and necessary measures at earlier stages.Machine learning(ML)techniques are used to assist healthcare providers in better diagnosing heart disease.This study employed three boosting algorithms,namely,gradient boost,XGBoost,and AdaBoost,to predict heart disease.The dataset contained heart disease-related clinical features and was sourced from the publicly available UCI ML repository.Exploratory data analysis is performed to find the characteristics of data samples about descriptive and inferential statistics.Specifically,it was carried out to identify and replace outliers using the interquartile range and detect and replace the missing values using the imputation method.Results were recorded before and after the data preprocessing techniques were applied.Out of all the algorithms,gradient boosting achieved the highest accuracy rate of 92.20%for the proposed model.The proposed model yielded better results with gradient boosting in terms of precision,recall,and f1-score.It attained better prediction performance than the existing works and can be used for other diseases that share common features using transfer learning.
基金supported by the National Natural Sci-ence Foundation of China(Nos.52375383 and 52035011).
文摘Electromagnetic sandwich metastructure(ESM)consisting of different functional layers,has gained in-creasing attention in radiation prevention and radar stealth.However,the current ESM design is primar-ily based on the separation design method,ignoring electromagnetic-mechanical interactions between layers.Thus,subject to thin thickness constraint of ESM,it is a great challenge to achieve broadband microwave absorption(MA)and excellent mechanical performance simultaneously.To address this is-sue,an electromagnetic-mechanical collaborative design approach was proposed for ESM.The relations of geometric-electromagnetic and geometric-mechanical of ESM were first identified by machine learn-ing.They were then integrated with the heuristic genetic optimization algorithm to perform the highly efficient design.The designed ESM can achieve 36.4 GHz effective absorption bandwidth(EAB,RL≤-10 dB),334.3 MPa equivalent bending strength and 83 MPa compressive strength with a thickness of 9.3 mm,possessing the widest EAB and highest bending strength within the current available MA struc-tures(thickness less than 9.5 mm).The proposed approach provides an efficient tool for the design of electromagnetic-mechanical optimal ESM.
文摘Low-voltage direct current(DC)microgrids have recently emerged as a promising and viable alternative to traditional alternating cur-rent(AC)microgrids,offering numerous advantages.Consequently,researchers are exploring the potential of DC microgrids across var-ious configurations.However,despite the sustainability and accuracy offered by DC microgrids,they pose various challenges when integrated into modern power distribution systems.Among these challenges,fault diagnosis holds significant importance.Rapid fault detection in DC microgrids is essential to maintain stability and ensure an uninterrupted power supply to critical loads.A primary chal-lenge is the lack of standards and guidelines for the protection and safety of DC microgrids,including fault detection,location,and clear-ing procedures for both grid-connected and islanded modes.In response,this study presents a brief overview of various approaches for protecting DC microgrids.
文摘The prediction of pregnancy-related hazards must be accurate and timely to safeguard mother and fetal health.This study aims to enhance risk prediction in pregnancywith a novel deep learningmodel based on a Long Short-Term Memory(LSTM)generator,designed to capture temporal relationships in cardiotocography(CTG)data.This methodology integrates CTG signals with demographic characteristics and utilizes preprocessing techniques such as noise reduction,normalization,and segmentation to create high-quality input for themodel.It uses convolutional layers to extract spatial information,followed by LSTM layers to model sequences for superior predictive performance.The overall results show that themodel is robust,with an accuracy of 91.5%,precision of 89.8%,recall of 90.4%,and F1-score of 90.1%that outperformed the corresponding baselinemodels,CNN(Convolutional Neural Network)and traditional RNN(Recurrent Neural Network),by 2.3%and 6.1%,respectively.Rather,the ability to detect pregnancy-related abnormalities has considerable therapeutic potential,with the possibility for focused treatments and individualized maternal healthcare approaches,the research team concluded.
