To address the challenges of high uncertainty, inter-subject variability, and inefficiency multi-feature utilization in motor imagery electroencephalogram (MI-EEG) classification, this study proposes amultiviewtransferTakagi-Sugeno-Kang (TSK) fuzzy classifier withsoftvariable embedded anddiscriminativestructural preservation (MVT-TSK-SVDS). First, a transfer learning mechanism incorporating soft variable embedding in the consequent part is developed. This mechanism establishes cross-domain correlations via a shared consequent and representation matrix. Within this framework, soft variable embedding and low-rank constrained discriminative learning work in concert to effectively capture supervision information and cross-domain relationships. Second, a local-global structural preservation term incorporating graph embedding and low-rank constraint is implemented to maintain local discriminative information from source domain while integrating global geometric patterns across all data. Third, a multiview adaptive learning framework is designed to address feature representation diversity and information loss during knowledge transfer. MVT-TSK-SVDS dynamically optimizes view-specific contributions through entropy maximization criterion while ensuring collaborative decision via consistency constraints. Experimental results validate strong generalization between and across datasets. Our model achieves 62.16% and 72.71% accuracy in cross-subject tasks on BCI-IV 2a and OpenBMI, respectively. In cross-dataset evaluations, it attains 62.75% accuracy on BCI-IV 2a to OpenBMI and 65.08% accuracy on OpenBMI to BCI-IV 2a, respectively.

Large Language Models (LLMs) are increasingly being applied in education, showing significant potential in personalized instruction, student feedback, and intelligent tutoring. Generating hints for Math Word Problems (MWPs) has become a critical application, particularly in helping students understand problem-solving steps and logic. However, existing models struggle to provide pedagogically sound guidance that fosters learning without offering direct answers. To address this issue, we introduce TMATH, a dataset specifically designed to evaluate LLMs’ ability to generate high-quality hints for MWPs. TMATH contains diverse mathematical problems paired with carefully crafted, human-generated hints. To assess its impact, we fine-tuned a series of 7B-scale language models using TMATH. Our results, based on quantitative evaluations and expert assessments, show that while LLMs still face challenges in complex reasoning, the TMATH dataset significantly enhances their ability to generate more accurate and contextually appropriate educational hints.

This paper proposes a creative, AI-driven integrated learning evaluation model that assesses metacognition and deeper learning through multimodal data analysis. Existing research faces challenges in learning data effectiveness, limited measuring tools and methods, and absence of feedback optimization loop. To address these issues, our ubiquitous multidimensional model integrate conscious and unconscious learning data using hybrid reasoning neural networks, generating interpretable representations aligned with metacognitive and deeper learning elements. This approach enables comprehensive assessment, automated feedback, and iterative optimization to enhance students' selfregulation and support students' personal learning needs. By advancing AI in Education (AIED), our integrated evaluation model explores new path for dynamic educational interventions and personalized pedagogies. This research contributes to the field by addressing validity issues, integrating qualitative and quantitative methods, and the loop with feedback optimization.

As the demand for precision, efficiency, and low-cost solutions in educational classification tasks continues to grow, enhancing model performance has become a critical focus of research. While large language models excel in these tasks, their high cost and resource requirements limit widespread application. This study proposes a Knowledge-Enhanced Distillation (KED) method, utilizing ChatGPT-4, ChatGPT-4o, and Llama3 as teacher models, and three different sizes of BERT models as student models. The method was validated across three real-world educational datasets. The results demonstrate that the KED method significantly improves the accuracy and F1 scores of small models in educational text classification tasks, while also substantially reducing computational costs and resource consumption. Notably, the KED method shows exceptional performance in scenarios involving few-shot learning and class imbalance. The innovation of this study lies in applying the KED method to educational classification tasks, filling a gap in current research and highlighting its significant potential for practical application in educational contexts.

Generative artificial intelligence (GAI) advancements have ignited new expectations for artificial intelligence (AI)-enabled educational transformations. Based on the theory of planned behavior (TPB), this study combines structural equation modeling and interviews to analyze the influencing factors of Chinese university students’ GAI technology usage intention. Regarding AI literacy, students’ cognitive literacy in AI ethics scored the highest (M = 5.740), while AI awareness literacy scored the lowest (M = 4.578). Students’ attitudes toward GAI significantly and positively influenced their usage intention, with the combined TPB framework and AI literacy explaining 59.3% of the variance. AI literacy and subjective norms positively influenced students’ attitudes toward GAI technology and perceived behavioral control, and attitude mediated the impact of AI literacy and subjective norms on GAI usage intention. Further, the interviews provide new insights for university management and educational leadership regarding the construction of an educational ecosystem under the application of GAI technology.

Faced with abundant online course resources, learners struggle to choose suitable materials. Learning resource recommendation algorithms can help address this. Rich online learning behavior data enables such recommendations. However, existing research only uses behavior events as learner features, ignoring event order i.e. Learning Behavior Patterns (LBPs). Also, only using click counts loses valuable information, hurting performance. We propose an algorithm leveraging online behavior sequences. First, extract sequences from logs and generate LBPs. Next, calculate Term Frequency Inverse Document Frequency (TF-IDF) values for each LBP as feature vectors. Cluster learners to improve efficiency. Finally, Calculate intra-cluster similarities for collaborative filtering recommendations. Experiments show over 30% precision, 9% recall, and 10% F1 improvements versus existing methods. Further ablation indicates learner clustering boosts time efficiency 3.75x without performance impact. Using TF-IDF values and tuning LBP length significantly improves performance. Overall, modeling orders via LBPs and better features like TF-IDF give major gains.

The emergence of educational video platforms has led to microlearning resources becoming increasingly mainstream. These platforms offer unique ecosystems and resource designs that better cater to the needs of learners. In this study, we examined the technology acceptance model (TAM) and task-technology fit (TTF) theory and conducted an empirical analysis of user satisfaction with new online learning spaces. We learned that perceived usefulness, perceived ease of use, and task-technology fit had significantly impacted user satisfaction, with these three factors collectively contributing to 78.2% of the variance in user satisfaction. Additionally, user satisfaction and task-technology fit significantly influenced the continuance intentions of users toward using these spaces, with both factors contributing to 66.7% of the variance in continuance intention. Overall, our findings revealed that the future development of new online learning spaces should consider the task requirements of learners and improve the platforms accordingly.

With the spread of COVID-19 in recent years, wearing masks has increased the difficulty of driver mental fatigue recognition. Electroencephalogram (EEG) signal has become an important physiological signal index to reflect the driver’s mental state. However, the drivers’ EEG data is plagued by inadequate labels and multi-view data, which makes classification difficult. To solve this problem, this study proposes asemi-supervisedmulti-viewsparse regularization andgraph embedding learning (SMSG) model. To obtain discriminative feature representations of semi-supervised EEG data, SMSG fully mines diverse information from multiple views based on sparse regularization embedding and graph embedding technology. SMSG employs the graph embedding to capture the discriminative structure and local manifold structure on multi-view data. Furthermore, SMSG learns the common shared regularization embedding and private regularization embedding factors to preserve the consistency and diversity of the multi-view data. Through self-adaptive learning, the weights of each view can be directly solved adaptively. This works also introduces kernel trick to project the SMSG model into the nonlinear reproducing kernel Hilbert space (RKHS), which can obtain more approximate EEG feature representation. Experiments on the real dataset verify the effectiveness of the SMSG model for EEG-based driver mental fatigue recognition.

With the development of knowledge economy globalization, problem-solving ability has become a necessary basic ability for students. Researches have confirmed that STEM teaching is an appropriate approach to improve students' problem-solving ability by taking interdisciplinary problems as the orientation. However, how to design STEM teaching experiment? What is the effect of STEM teaching in improving students' problem-solving ability? In order to answer the above two questions, this study adopted literature research method and quasi-experimental research method, and selected two classes of students in a middle school as research objects to carry out an empirical study for two months. Based on measurement tools such as problem-solving attitude questionnaire, problem solving-process observation table, and STEM work evaluation matrix, this study systematically and comprehensively verified the effect of STEM teaching on students' problem-solving ability. The results showed that STEM teaching has a positive effect on students' problem-solving attitude, problem-solving process and problem-solving results. Furthermore, the evaluation dimensions and tools for students' problem-solving ability in this study can be directly applied to relevant studies for future STEM teaching researches.

To evaluate how video-based reflection supported by the teacher dashboard influence teachers' abilities to reflect, 48 pre-service teachers from the same university in Eastern China were recruited to participating the comparative quasi-experiment to view the same classrooms video and make some discussion how to make the instructional design better in QQ. Twenty-four pre-service teachers were assigned in group in the experimental condition adopting video-based reflection with the teacher dashboard. Twenty-four pre-service teachers were the control condition adopting video-based reflection only. After analyze the teachers' collaborative discourse in conjunction with descriptive analysis alongside ENA, the results showed that video-based reflection supported by the teacher dashboard can improve the quality of reflection discussion and engage in deep and critical reflection to generate reason-based decisions with more teaching improvement.

Epilepsy recognition based on electroencephalogram (EEG) and artificial intelligence technology is the main tool of health analysis and diagnosis in Internet of medical things (IoMT). As a distributed learning framework, federated learning can train a shared model from multiple independent edge nodes using local data, which has greatly promoted the development of IoMT. One of the main challenges of EEG-based epilepsy recognition in IoMT is that EEG records show varying distributions in different devices, different times, and different people. This nonstationary characteristic of EEG reduces the accuracy of the recognition model. To improve the classification performance in IoMT, a hierarchical domain adaptation projective dictionary pair learning (HDA-PDPL) model is developed in the study. HDA-PDPL integrates EEG signals from different domains (person, edge nodes, devices, etc.) into a set of hierarchical subspace and simultaneously learns synthesis and analysis dictionary pairs in each layer. Specifically, a nonlinear transform function is introduced to seek hierarchical feature projection. The domain adaptation term on sparse coding builds a connection between different domains. Thus, the shared synthesis and analysis dictionaries can encode domain-invariant representation and discrimination knowledge from different domains. Besides, the local preserved term of projective codes is introduced to capture the potential discriminative local structures of samples. The experimental results on two EEG epilepsy classifications verified that the HDA-PDPL model can outperform other comparisons by utilizing more shared knowledge of different domains.

The safety monitoring system of intelligent transportation provides driving fatigue warning and risk control. Electroencephalogram (EEG) signals can directly reflect the neuronal activity of the brain. The detection and early warning of driving fatigue using EEG signals has important practical significance. However, because of the non-stationarity and timeliness of EEG signals, the single feature detection method is significantly impacted by data distribution differences. In this paper, in the framework of multi-input multi-output (MIMO) Takagi-Sugeno-Kang (TSK) fuzzy system, transferable TSK fuzzy classifier with multi-views (T-TSK-MV) is developed for EEG-based driving fatigue recognition in intelligent transportation. First, in view-specific consequent parameter learning, the view-specific consequent regularizer is designed based on technologies of ridge regression, maximum mean discrepancy (MMD), and manifold regularization, which becomes the bridge to transfer the discriminative information from the related domain to the target domain. In addition, the$\ell _{2,1} $-norm sparse constraint on consequent parameters is used to simplify fuzzy rules. Then multi-view learning is integrated into the consequent parameter learning, in which T-TSK-MV explores the view-shared consequent regularizer and adaptively assigns weights to each view. The$\ell _{2,1} $-norm sparse constraint on view-shared consequent regularizer can effectively exploit the local structure of multi-view data. Finally, the fuzzy classifier is constructed on view-specific regularizers and view weights. The experiment on real-word datasets shows that the proposed fuzzy classifier can significantly improve the driving fatigue recognition performance.

As a connection between teachers and students, instructional innovation behaviors can elicit students’ intrinsic motivation, facilitate them to engage in learning actively and deep learning, meanwhile the outcomes of instructional innovation can guide students to develop their capacities. To enhance teachers’ Al instructional creative performance, the study proposed an Al instructional creative performance model based on the organizational ecology and planned behavior theory, then inspected it by path analysis through a self-rating questionnaire survey of 496 K-12 teachers. The results showed that perceived organizational support and Al literacy do not have a direct effect on teachers’ Al instructional creative performance, but they have indirect impacts through the mediation effect of creativity intention and creative self-efficacy on that.

Abstract With the advancements in 5G network and mobile edge computing technology, autonomous vehicle technology has gained new development opportunities. Multitarget tracking becomes the research hotspots in autonomous vehicles. Since many factors such as motion blur, partial occlusion, and illumination changes affect the performance of target tracking, the problem of target tracking is still an open topic. In this article, inspired by the strong discriminative ability of dictionary learning, the discriminative dictionary learning with embedded auto‐encoder (DDLEA) algorithm is developed for the multitarget tracking system. The DDLEA algorithm integrates the auto‐encoder into the dictionary learning framework and learns sparse representations while preserving the local structure and discriminative information of data. The learned dictionary model has the strong recognition ability. Further, a multitarget tracking system is developed based on the proposed DDLEA algorithm and the hierarchical data association scheme. Based on the target confidence in the STKSVD model, the hierarchical data association method first uses the Hungarian algorithm to complete the preliminary matching of high confidence targets, and then further tracks the low confidence targets to improve the tracking ability. Experiments are carried out the public MOT 2015 dataset. Compared with several popular multitarget tracking algorithms, the tracking performance of our system is satisfactory.

Text sentiment classification is an important technology for natural language processing. A fuzzy system is a strong tool for processing imprecise or ambiguous data, and it can be used for text sentiment analysis. This article proposes a new formulation of a multi-task Takagi-Sugeno-Kang fuzzy system (TSK FS) modeling, which can be used for text sentiment image classification. Using a novel multi-task fuzzy c-means clustering algorithm, the common (public) information among all tasks and the individual (private) information for each task are extracted. The information about clustering, for example, cluster centers, can be used to learn the antecedent parameters of multi-task TSK fuzzy systems. With the common and individual antecedent parameters obtained, a corresponding multi-task learning mechanism for learning consequent parameters is devised. Accordingly, a multi-task fuzzy clustering–based multi-task TSK fuzzy system (MTFCM-MT-TSK-FS) is proposed. When the proposed model is built, the information conveyed by the fuzzy rules formed is two-fold, including (1) common fuzzy rules representing the inter-task correlation information and (2) individual fuzzy rules depicting the independent information of each task. The experimental results on several text sentiment datasets demonstrate the validity of the proposed model.

Cognitive computing is dedicated to researching a computing principle and method that can simulate the intelligence ability of human brain. Human emotion is the basic component of human cognitive activities. Electroencephalogram (EEG) computer signals obtained from a brain computer interface are difficult to conceal, and using machine learning methods to analyze EEG emotion is a hot topic in artificial intelligence. However, the EEG signal is non-stationary, making it difficult to select sufficient data from the same person to train a classifier for a subject. To promote the performance of emotion recognition methods, a multi-source domain transfer discriminative dictionary learning modeling (MDTDDL) is proposed in this study. The method integrates transfer learning and dictionary learning in a learning model, including the concepts of subspace learning, manifold smoothness, margin-based discriminant embedding, and large margin. The domain-specific transformation matrix projects EEG signals from various domains into the transfer subspace. The domain-invariant dictionary can find potential connections between multiple source domains and target domain. The manifold smoothness and margin-based discriminant embedding term further improve the model’s learning ability. The alternating optimization technique is used in model solving to efficiently compute model parameters. Experiments on the SEED and DEAP datasets demonstrate the effectiveness of MDTDDL.

Artificial intelligence– (AI) based fog/edge computing has become a promising paradigm for infectious disease. Various AI algorithms are embedded in cooperative fog/edge devices to construct medical Internet of Things environments, infectious disease forecast systems, smart health, and so on. However, these systems are usually done in isolation, which is called single-task learning. They do not consider the correlation and relationship between multiple/different tasks, so some common information in the model parameters or data characteristics is lost. In this study, each data center in fog/edge computing is considered as a task in the multi-task learning framework. In such a learning framework, a multi-task weighted Takagi-Sugeno-Kang (TSK) fuzzy system, called MW-TSKFS, is developed to forecast the trend of Coronavirus disease 2019 (COVID-19). MW-TSKFS provides a multi-task learning strategy for both antecedent and consequent parameters of fuzzy rules. First, a multi-task weighted fuzzy c-means clustering algorithm is developed for antecedent parameter learning, which extracts the public information among all tasks and the private information of each task. By sharing the public cluster centroid and public membership matrix, the differences of commonality and individuality can be further exploited. For consequent parameter learning of MW-TSKFS, a multi-task collaborative learning mechanism is developed based on ε-insensitive criterion and L2 norm penalty term, which can enhance the generalization and forecasting ability of the proposed fuzzy system. The experimental results on the real COVID-19 time series show that the forecasting tend model based on multi-task the weighted TSK fuzzy system has a high application value.

The purpose of this research was to explore how learners' non-cognitive abilities (including attention, engagement, presence, curiosity) differ in the learning process and the final learning results in VR and non-VR learning environments. This can be used to explain how the learning process of the learner in the VR environment occurs and changes, and the impact of the VR learning environment on the learner's non-cognitive ability. At the same time, it provided suggestions for the wider application of VR in education and the further development of personalized learning in education. Therefore, this research identified two research questions. The first one was whether the learning results of learners were different in VR and non-VR learning environments. The second was through the description of multimodal data such as eye movements and GSR. What was the relationship between the learner's non-cognitive abilities (attention, presence, engagement) and learning outcomes during the learning process? This study measured the curiosity of learners using a questionnaire before the experiment and measured the presence and engagement, compared scores between experimental group's and control groups' to understand how is immersed environment learning.

Classification of electroencephalogram (EEG) signal data plays a vital role in epilepsy detection. Recently sparse representation-based classification (SRC) methods have achieved the good performance in EEG signal automatic detection, by which the EEG signals are sparsely represented using a few active coefficients in the dictionary and classified according to the reconstruction criteria. However, most of SRC learn a linear dictionary for encoding, and cannot extract enough information and nonlinear relationship of data for classification. To solve this problem, a hierarchical discriminative sparse representation classification model (called HD-SRC) for EEG signal detection is proposed. Based on the framework of neural network, HD-SRC learns the hierarchical nonlinear transformation and maps the signal data into the nonlinear transformed space. Through incorporating this idea into label consistent K singular value decomposition (LC-KSVD) at the top layer of neural network, HD-SRC seeks discriminative representation together with dictionary, while minimizing errors of classification, reconstruction and discriminative sparse-code for pattern classification. By learning the hierarchical feature mapping and discriminative dictionary simultaneously, more discriminative information of data can be exploited. In the experiment the proposed model is evaluated on the Bonn EEG database, and the results show it obtains satisfactory classification performance in multiple EEG signal detection tasks.

Traditional clustering algorithms for medical image segmentation can only achieve satisfactory clustering performance under relatively ideal conditions, in which there is adequate data from the same distribution, and the data is rarely disturbed by noise or outliers. However, a sufficient amount of medical images with representative manual labels are often not available, because medical images are frequently acquired with different scanners (or different scan protocols) or polluted by various noises. Transfer learning improves learning in the target domain by leveraging knowledge from related domains. Given some target data, the performance of transfer learning is determined by the degree of relevance between the source and target domains. To achieve positive transfer and avoid negative transfer, a negative-transfer-resistant mechanism is proposed by computing the weight of transferred knowledge. Extracting a negative-transfer-resistant fuzzy clustering model with a shared cross-domain transfer latent space (called NTR-FC-SCT) is proposed by integrating negative-transfer-resistant and maximum mean discrepancy (MMD) into the framework of fuzzy c-means clustering. Experimental results show that the proposed NTR-FC-SCT model outperformed several traditional non-transfer and related transfer clustering algorithms.

Facial expression recognition has drawn increasing attention because of its great potential in human behavior analysis. Traditional recognition models usually assume that the training set is sufficient, and the training and testing data sets have the same distribution. However, these two factors are not satisfied in some cases. In this study, a transfer model collaborating metric learning and dictionary learning called TMMLDL is proposed to address the transfer facial expression recognition problem. To reduce the impact of cross-domain distribution variation, the information of global structure and pairwise constraints are utilized among training images in different domains. In particular, a discriminative metric space is learned into dictionary learning procedure such that the dictionary items can well present the discriminative information of different facial expression classes in the metric subspace. The proposed model tunes dictionary and metric space in an alternative optimization algorithm, which is guaranteed to obtain the optimal model parameters simultaneously. The experimental results on nine cross-domain facial expression classification tasks show that the proposed model achieves the satisfactory recognition performance.

Facial expression is one of the important characteristics of drivers during driving. It is very useful in safe driving detection. Recognizing drivers' expressions by the facial images can be solved with machine learning classification strategies. To obtain a reliable reorganization performance, most of approaches assume that the facial images in the training and testing datasets are independently and identically distributed. However, for real time drivers' facial expression recognition, due to vehicle motion, changes in illumination, noise and head movement, the features displayed for the training dataset may be not valid for the testing dataset. To solve this problem, a novel approach is proposed for cross-dataset transfer driver expression recognition via global discriminative and local structure knowledge exploitation in shared projection subspace (GD-LS-SS). By leaning a shared common subspace, GD-LS-SS utilizes the local geometrical structure of data by exploiting the knowledge of graph topology, meanwhile exploiting the global discriminative information by using the pairwise constrained knowledge between the source and labeled target data. Taking advantage of kernel trick, the kernel version of GD-LS-SS is proposed to learn the kernel projection for handling nonlinear cross-dataset transfer and to further promote the recognition accuracy. Experiments on the KMU-FED dataset show that the satisfactory recognition performance of GD-LS-SS outperforms several traditional non-transfer and related transfer approaches.

Morphological classification of human sperm heads is a key technology for diagnosing male infertility. Due to its sparse representation and learning capability, dictionary learning has shown remarkable performance in human sperm head classification. To promote the discriminability of the classification model, a novel local constraint and label embedding multi-layer dictionary learning model called LCLM-MDL is proposed in this study. Based on the multi-layer dictionary learning framework, two dictionaries are built on the basis of Laplacian regularized constraint and label embedding term in each layer, and the two dictionaries are approximated to each other as much as possible, so as to well exploit the nonlinear structure and discriminability features of the morphology of human sperm heads. In addition, to promote the robustness of the model, the asymmetric Huber loss is adopted in the last layer of LCLM-MDL, which approximates the misclassification error by using the absolute error function. Finally, the experimental results on HuSHeM dataset demonstrate the validity of the LCLM-MDL.

In this paper, a novel Bayesian Takagi-Sugeno- Kang (BTSK) fuzzy model and its joint learning method BTSK-JL of structure identification and parameter estimation are proposed for regression tasks from the perspective of Bayesian inference framework with a prior assumption about the number of fuzzy rules. Unlike most of existing TSK fuzzy systems where both their structure identification and parameter estimation of each fuzzy rule involved in them are learnt in a separate manner, BTSK-JL can determine the number of fuzzy rules and antecedent/consequent parameters of rules simultaneously in the proposed model. In order to guarantee their optimal solutions, BTSK-JL adopts a particle filter method to find the maximum-a-posterior value of the parameters. Due to taking into account the subtle interaction between input and output spaces, BTSK-JL can obtain good predictive performance and a set of compact fuzzy rules. Fuzziness and probability can work complementarily rather than competitively for such a TSK fuzzy system modeling. Experimental results on four time-series datasets and a glutamic acid fermentation process dataset have shown the validity and effectiveness of the proposed model.

As modern technologies have already become a part of our everyday life it is very important to clearly realize their impact on the educational process from the very first stages. That is why the current research compares the progress in English language learning among preschoolers in Ukraine who used technology-enhanced learning environment to those who were taught using traditional classrooms. After analyzing the data retrieved from 68 children and 5 teachers in Chernihiv private kindergarten (Ukraine), the findings indicate that the progress in general language proficiency, phonics skills and vocabulary knowledge of those who were taught with hand-held devices is much higher than that of those in traditional classroom settings. In addition, the results also reveal that modern technologies could help the preschoolers better focus on the materials, and engage them to participate more in the learning activities than paper-based materials.

A considerable body of researches indicate a high trend in researching social media. Accordingly, the study aimed to explore the relationship between social media relationships and academic self-concept through a correlation analysis. The data was collected from the 246 participants in secondary school who aged 12-16 years old from Shanghai, China. Results showed that all measures constructed for the study were internally consistent. Giving clues about social media, this study reveals that most students used blog, QQ, and WeChat frequently to connect with each other or share posts. The findings show there is a significant high positive relationship among positive peer relationships, academic persistence, academic expectation, learning interest, and academic self-concept.

Blended learning (B-Learning) is a common practice in higher education. The principles, characteristics, models, and strategies of B-learning have been widely investigated. B-Learning has also been used as a method for improving learning performance, achievement, motivation, interest, and problem-solving skills. However, the crucial differences between online learning and face-to-face sessions in terms of students' group interaction have not been fully explored. This study applied social network analysis and thematic analysis to investigate the nature and differences of the two environments. Group characteristic and interaction differences were analyzed using the 5090 posts and 604 dialogues of 53 participants engaged in a B-Learning course based on "Baidu Post Bar." Results show several differences in controlling pattern, depth, and content of interaction.

Customized content of e-textbook require teachers to spend greater efforts using authoring tools and planning activities before class, and teachers with various contexts have different demands on e-textbook. However, some teachers who lack ICT skills and dissatisfy with the features tend to give up using e-textbook. Thus we need to know the status of teachers' usage earlier before we decide to give them some technical supports. This paper describes an analysis method for predicting e-textbook adoption from usage records in early days, and an event segmentation method of teachers' usage is used in effort to provide features of predictive model.

In this paper, the Takagi-Sugeno-Kang (TSK) fuzzy classifier is casted into the Bayesian inference framework and a new fuzzy classifier called Bayesian TSK fuzzy classifier (B-TSK-FC) is proposed accordingly. The proposed classifier can be constructed by learning both the antecedent and consequent parameters of the involved fuzzy rules simultaneously. As a result of the introduction of Bayesian inference, the proposed B-TSK-FC can be distinguished as follows. 1) Unlike most existing TSK fuzzy classifiers where the antecedent and consequent parameters of fuzzy rules are learnt in a decoupled manner and the antecedent parameters are learnt only in the input space, the antecedent parameters of the fuzzy rules in B-TSK-FC are learnt by developing a fuzzy clustering method in the input-output space, and the consequent parameters of fuzzy rules are learnt in accordance with the maximum margin of separation principle, thereby resulting to form an intrinsic link between the input and output spaces to achieve improved classification performance and better interpretability. 2) With a Dirichlet prior assumption about fuzzy memberships in fuzzy clustering, a Markov-Chain Monte-Carlo technique is employed to estimate the parameters of the proposed classifier from a sampling perspective. 3) Rather than being rivals, fuzziness and probability in B-TSK-FC are collaboratively modeled to enhance the performance of TSK fuzzy classifier, in terms of classification and interpretability. Our experimental results in synthetic datasets as well as several real-world datasets confirm such merits of the proposed fuzzy classifier.

In this paper, a novel construction algorithm called imbalanced Takagi-Sugeno-Kang fuzzy classifier (IB-TSK-FC) for the TSK fuzzy classifier is presented to improve the classification performance and rule-based interpretability for imbalanced datasets. IB-TSK-FC consists of two components: 1) a cross-class Bayesian fuzzy clustering algorithm (BF3C) and 2) an imbalance learning algorithm. In order to achieve high interpretability, BF3C is developed to determine an appropriate number of fuzzy rules and identify antecedent parameters of fuzzy rules from the perspective of the probabilistic model. In addition to inheriting the distinctive advantage of Bayesian fuzzy clustering that the number of clusters can be estimated in the framework of Bayesian inference, BF3C considers repulsion forces between cluster centers belonging to different classes, and uses an alternating iterative strategy to obtain more interpretable antecedent parameters for imbalanced datasets. In order to improve the classification performance for imbalanced datasets, an imbalance learning algorithm is derived to estimate consequent parameters of fuzzy rules on the basis of the weighted average misclassification error. Comprehensive experiments on synthetic and UCI datasets demonstrate the effectiveness of the proposed IB-TSK-FC algorithm.

Video-based learning has gained popularity in higher education in recent years. Danmaku video is a kind of video where the screen is overlaid with user comments. In this study, the user comments consist of ideas and explanations about important concepts in the video, thus providing domain-specific knowledge and reducing the cognitive load for comprehension. This study tries to understand the effect of the danmaku video compared to with the normal video. Two groups of sophomore students (N= 20) were exposed to digital videos with or without danmaku. Both groups took part in a pre-and post-test on the topic of the given video. Time-locked eye movements were recorded to characterize participants' attention allocation to the Area of Interest (AOIs) consisting of danmaku, subtitles and teacher's face across the learning period. The results showed that danmaku video group outperformed the normal video group based on the increment of pre-and post-tests. Further, the percentage of fixation duration on each of the AOI was analyzed, and a significant difference was found in the amount of attention paid on different AOIs. The purpose of this study is focused on exploring the effects of Danmaku video in improving students' learning outcomes.

In this study, an educational decision supporting system regarding the issue of extending the length of Chinese national compulsory education policy is presented. In this system, the compulsory education funds are analyzed and calculated using the official Chinese statistical data. Integrated with the cloud computing technology, the system can forecast the school-age population and educational funds relative to the policy of extending the length of compulsory education. The educational data of Shanghai is selected as an example to test the validity and reliability of the proposed decision supporting system.

Previous research in term of traditional classroom has indicated that the effectiveness of English learning depends significantly on the strategy usage and different anxiety-provoking situations during communication. Meanwhile, researches in terms of technology-enhanced environment showed that the technology itself also can be a source of anxiety. This paper aims to understand the correlation between LLS usage and two types of anxiety – communication and technology. We adopted the FLASC and SILL questionnaires in order to collect the data from 187 Chinese University Students. The results showed that learners more often used direct LLS than the indirect ones; the most popular are compensation strategies and the least - affective. In addition, findings indicated that there was no significant difference between technology anxiety and the LLS usage. However, the communication anxiety influenced significantly the choice of LLS, especially social, cognitive, affective and memory strategies.

Today, the fields of international education, economic and politics exchanges closely, with the patterns of cultures exchanging closely. Settled into the game, in the intercultural era, how are the difference behind the educational games design under different culture context? On this issue, this research makes intercultural value comparisons between the US and China, and have an inspection of design mechanisms about cultural impact on educational game from the Hofstede Cultural Dimensions. This research adopt a comparative study of currently educational game design between China and the U S, aimed at identifying cultural differences between the games with a view to provide a basis for design innovation for the educational games in cultural diversity.

Mobile learning is the ongoing project that aims at fulfilling learning requirements in informal settings for Shanghai citizens. In order to understand whether users like using the mobile learning materials, an empirical study was conducted in a simulated field situation that was based on an integrated model. Fifty volunteers were recruited from the target population to participate in this study. The results showed that perceived quality had a significant influence on user satisfaction, as well as on individual intention to adopt. The findings of this study will help mobile learning content providers to develop learning materials based on an understanding of the perceptions of potential adopters.

Recently, ISO/JTC1 SC36 has initiated a new standard project on e-Textbook, which was proposed by the authors’ team. In this project, the information model of e-Textbook is the most fundamental part and further studies all will be based on it. The purpose of this paper is to manifest the information model of the e-Textbook by describing a practical case in our ongoing work. The process of designing an English e-Textbook for grade 6 is presented to explain the main steps of element structure design and function structure design. The information model and its practice reflect what kind of e-Textbook we need and how to design a standard e-Textbook for discussing the disorder of e-Textbook’s development.