Search

The method of computational intelligence to monitor and evaluate the concentration of students in the teaching process can promptly and effectively adjust the learning plan and improve the learning effect. In this article, clustering algorithm and fuzzy control methods are used to construct a research model of students’ attention in class. In addition, this article uses the existing MATLAB-based image feature recognition algorithm to detect and obtain facial features and analyze the main features of facial expressions through computational techniques to realize the judgment of attention. In addition, this article optimizes the traditional AdaBoost algorithm to save computing time and improve operating efficiency and system performance stability. Finally, this article constructs the functional modules of the research model according to actual needs and designs experiments to verify the performance of the model. Experimental research results show that the model constructed in this article has a certain effect.

Introduction: The COVID-19 pandemic accelerated use of technology like videoconferencing (VC) in healthcare settings to maintain clinical teaching and continuous professional development (CPD) activities. Sociomaterial theory highlights the relationship of humans with sociomaterial forces, including technology. We used sociomaterial framing to review effect on CPD learning outcomes of morbidity and mortality meetings (M&M) when changed from face-to-face (FTF) to VC. Methods: All surgical department staff were invited to participate in a survey about their experience of VC M&M compared to FTF M&M. Survey questions focused on technological impact of the learning environment and CPD outcomes. Respondents used 5-point Likert scale and free text for qualitative responses. De-identified data was analysed using Chi-squared comparative analysis with p

Sports psychology is a branch of psychology and is a sports course that takes both theory and practice into account. It is not only an integral part of the talent training curriculum system of sports colleges and universities, but also a basic compulsory course for social sports majors in comprehensive universities. Sports psychology has currently become one of the significant theories in PE teaching, which not only guides PE teaching, but also helps solve existing problems in PE teaching and drives the development of PE teaching. Subject to the impact of traditional education mode and educational concept, etc., however, there are some problems in the existing the teaching of sports psychology, such as students’ low interest, poor initiative and poor educational effect. To raise students’ learning interest, this paper selected a problem-oriented modular teaching mode, and applied it to the sports psychology classroom, to improve the learning effect. In this study, we first used data mining technology to build an anxiety prediction model and then identified a problem to be solved. After that, the problem was designed and analyzed, the known problem was confirmed, new information was collected, old and new knowledge was integrated, until the problem was finally solved. On this basis, based on virtual simulation technology, an experimental simulation teaching system for sports psychology was presented. It was found through research that the problems-oriented teaching mode can significantly enhance the effect of sports psychology and PE teaching and increase students’ learning initiative. The learning effect was significantly better than that of traditional teaching methods.

The cultivation of intellectual property right talents who bear the mission of blockchain era becomes the key in the new breakthrough of integrating blockchain and intellectual property rights to form new sources of economic growth in the new/old kinetic energy conversion. Higher education is the major way to cultivate talents. Nevertheless, blockchain law courses currently offered by international colleges and universities are developed from technology fields of technology finance, distributed systems, and cryptocurrency, but not the instruction from legal perspectives. The experimental design model is applied in this study to the experimental research of students majoring in intellectual property rights in the same grade from same-level schools in Guangxi. Total 170 students are preceded the 16-week blockchain law knowledge on-campus experimental teaching. The research results show significant correlations between blockchain law course integrated multimedia dynamic teaching and learning motivation, remarkable correlations between blockchain law course integrated multimedia dynamic teaching and learning effectiveness, notably positive effects of learning motivation on learning effect in learning effectiveness, and significantly positive effects of learning motivation on learning gain in learning effectiveness. According to the results, suggestions are proposed, expecting to cultivate domestic students majoring in intellectual property rights, from legal vision, establishing the systematic way of thinking and knowledge system from know-why, to practical application, and then to legal regulations to enhance the learning effect of blockchain law knowledge and core skills to solve blockchain law issues of students majoring in intellectual property rights.

With the rapid development of network technology, computer viruses have developed at a fast pace. The threat of computer viruses persists because of the constant demand for computers and networks. When a computer virus infects a facility, the virus seeks to invade other facilities in the network by exploiting the convenience of the network protocol and the high connectivity of the network. Hence, there is an increasing need for accurate calculation of the probability of computer-virus-infected areas for developing corresponding strategies, for example, based on the possible virus-infected areas, to interrupt the relevant connections between the uninfected and infected computers in time. The spread of the computer virus forms a scale-free network whose node degree follows the power rule. A novel algorithm based on the binary-addition tree algorithm (BAT) is proposed to effectively predict the spread of computer viruses. The proposed BAT utilizes the probability derived from PageRank from the scale-free network together with the consideration of state vectors with both the temporal and learning effects. The performance of the proposed algorithm was verified via numerous experiments.

The development of educational information technology makes the construction and application of online teaching space more and more common. As a field and assistant system, the essential requirement of online-teaching space is to ensure that learners can continuously engage in learning. The goal of the application of flow theory in teaching is to stimulate learners' flow experience, so as to improve the learning effect and achieve the learning goal. Starting from the design of online teaching based on flow theory, this paper puts forward the design requirements and principles of its teaching links, including the analysis of individual chemical situation, hierarchical and classified teaching objectives and tasks, multi scene teaching links and teaching improvement, and then puts forward the corresponding online teaching space framework. This paper argues that online teaching space should be composed of intelligent guidance system, teaching support system, situational intervention system and flow support system. Under such conditions, teaching space is more conducive to the implementation of online instructional design based on flow theory, and can help space constructors and participants achieve continuous improvement of teaching effect.

Transfer learning methods are conventionally conducted by utilizing abundant labeled data in the source domain to build an accurate classifier for the target domain with scarce labeled data. However, most current transfer learning methods assume that all the source data are relevant to target domain, which may induce negative learning effect when the assumption becomes invalid as in many practical scenarios. To tackle this issue, the key is to accurately and quickly select the correlated source data and the corresponding weights. In this paper, we make use of the least square-support vector machine (LS-SVM) framework for identifying the correlated data and their weights from source domain. By keeping the consistency between the distributions of the classification errors of both the source and target domains, we first propose the classification-error-based consensus regularization (CCR), which can guarantee the performance improvement of the target classifier. Based on this approach, a novel CCR-based selective transfer classification learning method (CSTL) is then developed to autonomously and quickly choose the correlated source data and their weights to exploit the transferred knowledge by solving the LS-SVM based objective function. This method minimizes the leave-one-out cross-validation error despite scarce target training data. The advantages of the CSTL are demonstrated by evaluating its performance on public image and text datasets and comparing it with that of the state-of-the-art transfer learning methods.

This paper firstly designs a five-dimensional model of learners’ characteristics (learners’ English reading ability, cognitive style, learning goal, learning situation, and learning effect) and a three-dimensional model of English reading resources’ characteristics (question types, topics, and difficulty of resources) in a fragmented learning environment through literature research. At the same time, to make the learning resources meet the characteristics of fragmented learning time and space, the English Level 4 reading resources are reasonably designed and segmented to adapt to the needs of learners’ mobile fragmented learning. Then, combined with machine learning algorithms, an adaptive recommendation model of learning resources in English fragmented reading is constructed. The algorithm-based adaptive recommendation algorithm for English fragmented reading resources is designed. Based on the generated decision trees, the expression rules are parsed to achieve adaptive pushing of resources. The results of this study show that adaptive recommendation of learning resources in English fragmented reading can help teachers to develop future resource recommendation strategies through effective data collection to adaptively push resources that are close to learners’ individual needs. The use of mobile by English learners to learn to read in a fragmented learning context enables targeted training in weak areas of English reading, thus enhancing different aspects of learners’ reading skills.

With the rapid development of network technology and multimedia technology, especially in the special environment of 2020, online learning has become one of the main learning methods under the digital learning environment. At the same time, in the process of constructing digital learning resources, online learners are prone to new problems such as loss of focus and fluctuation of learning effect. How to fundamentally stimulate and maintain the learning interest of online learners in the process of constructing digital learning resources has become an urgent problem to be solved. Based on the problem-oriented approach and the comprehensive application of in-depth interview method and other research methods, this paper finds out four major problems affecting learning interest based on the in-depth analysis of the basic characteristics and learning status of online learning in the construction of digital learning resources. This paper analyzes the reasons and puts forward solutions from three aspects: scientific design and development of resources, active and effective incentive strategies, and rich resource construction paths, so as to provide useful references for stimulating and maintaining online learners’ interest in learning in the construction of digital learning resources.

The current life-prediction models for lithium-ion batteries have several problems, such as the construction of complex feature structures, a high number of feature dimensions, and inaccurate prediction results. To overcome these problems, this paper proposes a deep-learning model combining an autoencoder network and a long short-term memory network. First, this model applies the characteristics of the autoencoder to reduce the dimensionality of the high-dimensional features extracted from the battery data set and realize the fusion of complex time-domain features, which overcomes the problems of redundant model information and low computational efficiency. This model then uses a long short-term memory network that is sensitive to time-series data to solve the long-path dependence problem in the prediction of battery life. Lastly, the attention mechanism is used to give greater weight to features that have a greater impact on the target value, which enhances the learning effect of the model on the long input sequence. To verify the efficacy of the proposed model, this paper uses NASA’s lithium-ion battery cycle life data set.

Digital health metrics promise to advance the understanding of impaired body functions, for example in neurological disorders. However, their clinical integration is challenged by an insufficient validation of the many existing and often abstract metrics. Here, we propose a data-driven framework to select and validate a clinically relevant core set of digital health metrics extracted from a technology-aided assessment. As an exemplary use-case, the framework is applied to the Virtual Peg Insertion Test (VPIT), a technology-aided assessment of upper limb sensorimotor impairments. The framework builds on a use-case-specific pathophysiological motivation of metrics, models demographic confounds, and evaluates the most important clinimetric properties (discriminant validity, structural validity, reliability, measurement error, learning effects). Applied to 77 metrics of the VPIT collected from 120 neurologically intact and 89 affected individuals, the framework allowed selecting 10 clinically relevant core metrics. These assessed the severity of multiple sensorimotor impairments in a valid, reliable, and informative manner. These metrics provided added clinical value by detecting impairments in neurological subjects that did not show any deficits according to conventional scales, and by covering sensorimotor impairments of the arm and hand with a single assessment. The proposed framework provides a transparent, step-by-step selection procedure based on clinically relevant evidence. This creates an interesting alternative to established selection algorithms that optimize mathematical loss functions and are not always intuitive to retrace. This could help addressing the insufficient clinical integration of digital health metrics. For the VPIT, it allowed establishing validated core metrics, paving the way for their integration into neurorehabilitation trials., npj Digital Medicine, 3 (1), ISSN:2398-6352

At present, the study of personalized student models in large-scale open learning environments often selects the characteristic dimensions of students based on human experience, and there is a lack of research on whether student characteristics are effective to learning behavior analysis or whether it can affect learning effects. It is necessary to explore the impact of learning behavior characteristics on learning effects through data mining, extracting multi-dimensional features that can comprehensively and accurately reflect students’ abilities and preferences, to build a student model suitable for personalized strategy recommendations and resource recommendations. The paper uses log data of online learning platforms to explore the impact of different behavioral dimensions on students’ learning enthusiasm using correlation analysis, independent sample testing and other methods. By discovering the characteristics of different types of students, combining evaluation indicators and improved clustering algorithms, characteristics of 4 groups with different learning strategies are creatively proposed.

Although VR technology can provide an ideal learning and application environment for learners’ language skill acquisition, the learning performance of different types of learners in virtual environments and differences in their knowledge transfer ability from the virtual to the real environment still need further discussion. Therefore, we developed a VR English speaking training and testing system to understand the influence of cognitive style and test environment on the learners’ learning effect. The results indicated that: (1) the learning effect of the field-independent learners was lower than that of the field-dependent learners in the real testing environment, but significantly higher than that of the field-dependent learners in the virtual testing environment. Meanwhile, there was a more significant difference in the real and virtual learning effect between the field-dependent and field-independent learners, (2) there was a significant interaction between cognitive style and test environment in the learners’ learning effect. Besides, cognitive style and test environment had an influence on the spoken English learning effect based on VR. The field-independent learners were more likely to transfer what they had learned in the virtual environment to the real application.

With the rapid development of big data era, there are more and more research and application of data-based mining and analysis technology in education. This paper discusses how to use data mining technology to integrate learning data into college English classroom, and to track and analyze students’ learning situation dynamically through data mining in student-centered college English teaching. According to the constructed academic diagnosis model, accurately diagnose learning problems, accurately locate all kinds of learning groups in the classroom, correctly locate learning needs, realize the push of personalized learning resources, and promote the improvement of learning effect. Through classified and hierarchical teaching intervention, we can guide learners to cultivate their interest in learning and improve teachers' grasp of the classroom and the efficiency of teaching. Create a positive and effective learning community.

Machine learning has been proved to be feasible and reasonable in architectural field by extensive researches recently, whereas its potential is far from being tapped. Previous studies show that the training of GAN by labelling can enable a computer to grasp interrelationship of spatial elements and logical relationship between spatial elements and boundary. This study set the learning object as layout of private gardens in southern Yangtze with higher complexity. Chinese scholars usually analyse private garden layout based on their observation and experience. In this paper, based on Pix2Pix model, we enable a computer to generate private garden layout plan for given site conditions by learning classic cases of traditional Chinese private gardens. Through the experiment, taking Lingering garden as example, we continuously adjust the labelling method to improve learning effect. The finally trained model can quickly generate private garden layout and aid designers to complete scheme design with private garden element corpus. In addition, the working process of training GAN enables us to discover and verify some private garden layout rules that have not been paid attention to.

In post-COVID-19 pandemic era, students'' learning effects and experience may sharply decrease when teaching is transferred from offline to online. Several tools suitable for online teaching have been developed to guarantee and promote students’ learning effects. However, they cannot fully consider teacher-student interaction in online teaching. To figure out this issue, this paper proposes integrated media platform-based virtual office hours implementation for online teaching. Specifically, an integrated media platform (IMP) is first constructed. Then, virtual office hours (VOH) is implemented based on the IMP, aiming at increasing student-teacher interactions. For evaluating the effectiveness of this scheme, 140 undergraduate students using IMP are divided into one control group and three experimental groups that respectively contain text, voice and video modes. The experiment results indicate that applying VOH in the IMP can improve students’ online presence and test scores. Furthermore, students'' participating modes during VOH implementation can largely affect their degree of presence, which can be well classified by using principal component analysis. The implication of this work is that IMP-based VOH is an effective and sustainable tool to be continuously implemented even when the COVID-19 pandemic period ends.

With the continuous development of the times, intelligent teaching assistance is also attracting more and more attention in education. The intelligent detection algorithm for student behavior is gradually becoming more precise. In university classrooms, students sleep on mobile phones and are more serious. Intelligent education can more accurately identify student behaviors to help teachers optimize teaching methods, thereby improving students' classroom learning effects. This paper studies and improves YOLOv4, and proposes a network structure called YOLOv4-Bi, which mainly adds the enhanced feature extraction network of YOLOv4 to the feature extraction structure of jumping and top-down, bottom-up combined paths. The used student classroom recording video is enhanced by taking the frame data and training, and testing on this data set. The original YOLOv4 is compared with the network of the improved PANet module and the Faster R-CNN network, and the data is carried out in the data set. It is verified that the mAP of the improved YOLOv4 network is higher than the mAP of the original unimproved YOLOv4 network. Compared with the original network, YOLOv4 is more suitable for student detection and recognition.

Hybrid teaching, a typical result of the deep integration of information technology and education, has become increasingly normalized in higher education in the COVID-19 epidemic and post-epidemic era. The teaching model integrates the advantages of online teaching and traditional teaching. However, if online resources are monotonous and offline practices are not well connected with online instruction, the hybrid teaching reforming will only stay on the surface, and the learning effect will decrease rather than improve. Resources and activities are two critical issues to achieve effective hybrid teaching. Specifically, the first is how to construct an online learning resource system suitable for the development of offline teaching, and the second is how to properly use online resources to carry out offline learning activities. This paper proposes a hybrid teaching mode with students as the first perspective by improving the BOPPPS model. The teaching mode makes the combination of online and offline teaching throughout the three stages of pre-class, in-class and after-class, and adopts three closed-loop feedback and three-level participatory interaction. The whole teaching process reflects the dominant position of students in teaching resources, teaching methods, teaching activities, teaching evaluation and so on. The teaching mode can realize the transformation from teacher-centered to student-centered in the teaching process, and realize the transformation of teaching achievement from learning knowledge to having ability.

Bugs in learners' programs are often the result of fundamental misconceptions. Teachers frequently face the challenge of first having to understand such bugs, and then suggest ways to fix them. In order to enable teachers to do so effectively and efficiently, it is desirable to support them in recognising and fixing bugs. Misconceptions often lead to recurring patterns of similar bugs, enabling automated tools to provide this support in terms of hints on occurrences of common bug patterns. In this paper, we investigate to what extent the hints improve the effectiveness and efficiency of teachers in debugging learners' programs using a cohort of 163 primary school teachers in training, tasked to correct buggy Scratch programs, with and without hints on bug patterns. Our experiment suggests that automatically generated hints can reduce the effort of finding and fixing bugs from 8.66 to 5.24 minutes, while increasing the effectiveness by 34% more correct solutions. While this improvement is convincing, arguably teachers in training might first need to learn debugging "the hard way" to not miss the opportunity to learn by relying on tools. We therefore investigate whether the use of hints during training affects their ability to recognise and fix bugs without hints. Our experiment provides no significant evidence that either learning to debug with hints or learning to debug "the hard way" leads to better learning effects. Overall, this suggests that bug patterns might be a useful concept to include in the curriculum for teachers in training, while tool-support to recognise these patterns is desirable for teachers in practice., To be published in the proceedings of the 16th Workshop in Primary and Secondary Computing Education (WIPSCE)

Modern society needs to think of new approaches for solving problems with computing. Computational thinking is the process of abstracting and automating a variety of problems using computational technology. A system that expresses, manages, and processes knowledge such as computational thinking is called a knowledge-based system. This paper proposes to examine students’ knowledge about computational thinking when they want to develop a Python project, and the correlation/association between these concepts. To achieve our goal, a field study was designed and data were collected from a computer programming lecture. Through this data analysis, we try to identify the factors through the correlation between data and clustering technique in order to express and discover the knowledge about the learner’s computational thinking. For the verification of the factors identified, we analyzed the correlation between computational thinking and the pre- and post-test results of the LightBot. In addition to the regression analysis of the proven factors, the probability of the research model was analyzed through the structural equation to process the knowledge discovered. In this paper, we present various problems in the domain of programming education and analyze the means to diagnose and improve knowledge based on computational thinking by finding various problem-solving methods. To pre-examine the learner; he/she was diagnosed using a test paper and the LightBot execution test. We checked the learner’s current knowledge state by analyzing the correlation between the test site and the results of the LightBot. To analyze the level of knowledge improvement of learners, we designed an experiment to analyze the correlation between learning and the actual test results through a system that applied the problem-solving learning method. An analysis of the experimental results demonstrated that there was a correlation between the test results for a learner and the pre-test results of the LightBot. Additionally, the group mean scores of the learners who learned as per the proposed technique were observed to be significant. During this process, we analyzed the effects of problem-solving and system application on academic achievement through factor analysis, regression analysis, and structural equation modeling. The ability to pinpoint various problem scenarios and solve problems more effectively using computational technologies will become more important in future. For this purpose, applying our proposed technique for deriving and improving knowledge based on computational thinking to software education will induce the interest of students and increase the learning effect.

The influence of learning effects on job scheduling has been studied for years. By considering learning effects, an operator can schedule jobs in a more precise way and improve on the original sche...

The objective of thinking development lesson is to help students change and develop thoughts via classroom teaching. In thinking development lessons, students will reinforce the application scenarios of their thinking skills and learn new thinking skills. For design of classroom teaching of thinking development type, thinking visualization tools are needed for organizing and guiding thinking process. In this study, a teaching design framework of thinking development lesson was developed based on mind mapping, and applied in the teaching design process of the course of Health Fitness Education. It consists of five steps, namely teaching objective setting, question situation establishment, externalization of tacit thinking, formalization of explicit thinking, and automation of efficient thinking. Teaching result check shows that the course teaching design scheme developed in this study is helpful to improve students’ learning effect in the course of Health Fitness Education significantly. This study provides a new teaching design method for teaching of Health Fitness Education, which is of important significance to helping students integrate the knowledge they have learned and rationalize thinking.

Along with the rapid development of information technology, the information technology education trend is emerged. The enhancement of quality becomes the cross-century direction of education reform. To grasp the trend of time and integrate information multimedia into traditional teaching courses for enhancing teaching quality are the primary issues for professional teachers. With an experimental design model, Total 100 students of Hubei University of Automotive Technology, as the research objects. The class is an 8-week experimental instruction every Thursday (32 hours in total, 16 hours in class, another 16 hours are for out-of-class activities). The research results reveal that information multimedia instruction shows higher self-efficacy than general instruction and achieves the significance, information multimedia instruction reveals higher learning effect than general instruction and reaches the significance, and self-efficacy appears notably positive effects on learning effect. According to the results to propose suggestions, it is expected, through the integration of information multimedia, to induce and maintain students’ interests in the curricula and even the willingness to make efforts towards the learning goal for better learning effect.

Academic emotions can produce a great impact on the learning effect. Normally, emotions are expressed externally in the students' facial expressions, speech and behaviour. In this paper, the focus is on automatic academic emotion inference based on facial expressions in online learning. Considering the lack of training samples for the inference algorithm, a spontaneous facial expression database is established. It includes the facial expressions of five common academic emotions and consists of two subsets: a video clip database and an image database. A total of 1,274 video clips and 30,184 images from 82 students are included in the database. The samples are labelled by both the participants and external coders. An extensive analysis is carried out on the image database using a convolutional neural network (CNN)-based algorithm to infer self-annotation. Some data augmentation algorithms are applied to improve the algorithm performance. Additionally, an adaptive data augmentation algorithm based on spatial transformer network is introduced, which can remove some confounding factors in the original images. The algorithm can obviously improve the inference performance, which has been proven by comparing some evaluation indicators before and after adoption. Such a database will certainly accelerate the application of affective computing in the educational field.

Turtle graphics, first introduced in the Logo programming language in the late 1960s, have since been widely used in programming education. Turtle graphics is excellent for programming training, providing an intuitive platform for understanding the fundamental procedural programming principles, through its simple graphics functions allowing visual feedback for programming. This paper introduces a new and original application of turtle graphics to Java programming language, particularly targeted to early Java programming education. The aim of the proposed Java language turtle graphics library is (a) to use turtle graphics to provide intuitive understanding of fundamental procedural programming principles, (b) to provide an easy tool for the students to create visually interesting programs in the early stages of programming education, and (c) to stimulate the motivation of students to improve the learning effect of programming training and encourage further self-study. The proposed Java language turtle graphics library was introduced to 1st year college students with minimal Java language programming training, to verify the effectiveness of the proposed method.

A three-dimensional visual flight control simulation experiment platform based on FlightGear is designed to study the flight control of quadrotors. The simulation model and flight control algorithm are written in Matlab/Simulink. The platform’s functions include real-time display of three-dimensional flight scenes, display of flight control and corresponding processes, and display of curves of flight status variables. The simulation platform intuitively shows the correlation between mathematical variables and their physical meanings, and promotes students' understanding of control theory. In addition, the platform also has a strong openness, which can meet the needs of students to independently learn flight control theory and design control algorithms, and effectively improve the learning effect.

Popular science micro video is a form of popular science with the aim of knowledge popularization. It usually lasts for 30 seconds to 20 minutes and mainly was spread on mobile terminals. Often it contains a lot of interesting, attractive but non-important content, and the content may not directly relate to the main knowledge points of popular science. Such content can be called seductive details. Studies have shown that adding seductive details to teaching materials will have negative impacts on learning. However, popular science micro video is a form of multimedia with low cognitive difficulties and constant changes in media and learning environment. In addition, the effect of seductive details may change. The purpose of this study is to examine the influence of seductive details in popular science micro videos on learning effects. In this experiment, 44 students were selected to watch popular science micro videos that have seductive details. Then the relationships between the memory amount of seductive details, accumulation of prior knowledge and the learning effect of science knowledge (memorizing and transferring) were analyzed. According to the results, in the case of low cognitive load, the addition of seductive details in popular science micro videos had no effect on the memorizing effect of popular science knowledge, but improved the transferring effect. The learning effect of popular science knowledge had nothing to do with audience's prior knowledge level. Therefore, this study suggested that some seductive details should be added to popular science micro videos but the difficulty of making popular science micro videos should be reasonably controlled, which would make the videos more interesting and help enhance the learning effect of popular science knowledge.

The change of social demand for English application-oriented talents has affected the transformation of the way of cultivating talents in College English. The traditional training mode of indoctrination has exposed more and more disadvantages, and the demand of new teaching mode is imminent. This paper first analyzes the current situation of College English learning platform in China. Then, it discusses the design and implementation of “Online + offline” College English learning platform based on Web. The overall structure and function of the database are designed in detail. The teaching platform provides students with learning tools, learning resources, communication platform, testing and evaluation functions, and can evaluate students’ learning behavior, learning process and learning effect. This paper traces and collects a large amount of data left by learners in learning college English courses, and analyzes learners’ learning habits, learning progress and learning effect. Finally, according to the learning big data, this paper customized a reasonable personalized learning platform and improved the online teaching personalized service system.

BackgroundCancer risk calculators were introduced to clinical practice in the last decade, but they remain underused. We aimed to test their potential to improve risk assessment and 2-week-wait referral decisions.Methods157 GPs were presented with 23 vignettes describing patients with possible colorectal cancer symptoms. GPs gave their intuitive risk estimate and inclination to refer. They then saw the risk score of an algorithm (QCancer was not named) and could update their responses. Half of the sample was given information about the algorithm’s derivation, validation, and accuracy. At the end, we measured their algorithm disposition.ResultsGPs changed their inclination to refer 26% of the time and switched decisions entirely 3% of the time. Post-algorithm decisions improved significantly vis-à-vis the 3% NICE threshold (OR 1.45 [1.27, 1.65], pConclusionsCancer risk calculators have the potential to improve 2-week-wait referral decisions. Their use as learning tools to improve intuitive risk estimates is promising and should be further investigated.

At present, the new teaching mode based on the "online + offline" combination of MOOC teaching and the interesting flipped classroom has exerted a strong influence on the adoption of traditional mathematics teaching methods in colleges and universities. The current mathematics teaching especially urgently needs to solve the difficult problem of university mathematic teaching in colleges and universities, is to design a mathematical intelligent computer assisted instruction (ICAI) MOOC platform, the platform can according to students' actual learning effect, intelligent meet the demand of the characteristics of students, can for the teachers and students interact through man-machine interaction experience, thus to manage the process of teachers' teaching to the student, to evaluate students' learning effect. To meet these requirements, based on BP neural network algorithm and using SSH framework, this topic designs a SET of MOOC platform to facilitate students to learn mathematics.

In order to better predict the learning effect of blended learning, provide early warning and intervention of learning, and improve the teaching quality, this paper proposes a model based on RRF(Rule and Random Forest), which can be used to classify the posts on course forum, and the classification results are calculated to obtain the score of course forum posts. Then, the SVM (Support Vector Machine) model fused with posts classification is used to predict the students’ grades. After adding the scores of forum posts, the prediction accuracy increased from 85.9% to 89.03%. This learning effect prediction model was applied in blending learning, and successfully reduced the failure rate in prediction of students’ final grades.

With the rapid development of information technology, the integration of systematic learning of online education platform and fragmented learning of mobile terminal has given birth to a new hybrid learning mode. The key to the integration of the two platforms is the effective integration of learning resources, so that all kinds of resources complement each other. In order to adapt to the change of learners’ learning style, and promote the learning effect of learners. To integrate the resources of the two platforms, we should follow the principles of systematization, miniaturization of knowledge points, simplicity of presentation and through use. This paper analyzes the structure and content of online education platform learning resources, establishes curriculum learning resource database, selects the appropriate learning resources of mobile learning platform (Superstar Learning APP as an example), selects the resource presentation mode suitable for mobile terminal, and designs mobile learning resources based on new hybrid learning.

Computer practice teaching need efficient and lightweight online collaborative learning platform between teachers and students. Based on CSCL learning scenario, we propose a method of building online teaching platform of computer practice course by combining various lightweight mobile Internet tools, and conducts an empirical study in the course of “Innovation Engineering Practice”. The results show that lightweight online collaboration platform can promote efficient interaction between learners, improve self-management of learners, and improve learning effect and self-evaluation of learners.

The accurate teaching mode is the long-term education practice in our country to create "according to their aptitude" the principle of integration with modern education technology, through the measurement process to track the students' academic performance and learning effects, according to different learners, design different target, selection of different teaching contents, teaching to provide data for subsequent reference, accurate guide teaching.

The application of digital resources is very important in the design of English teaching platform. Traditional English teaching resources can not be matched precisely because of the narrow source of digital teaching resources. Web mining technology makes personalized teaching possible. The web mining technology to track and collect a large amount of data left by learners in the online course of modern English grammar is used in this paper. Through the analysis of the learners' learning habits, learning progress, learning effect and so on, we can customize a reasonable personalized learning platform for them. The experimental results show that the platform can improve the personalized online teaching service system, optimize teaching resources, and build a customized online teaching platform.

Discrete mathematics is the basic course of computer major. This paper aims at the present situation of discrete mathematics teaching, the design scheme of course content, teaching process and course examination based on fine examination is put forward. And through the CG platform, students' pre class test, homework submission, stage test, platform exchange and interaction and other teaching processes are effectively managed. Through the data analysis of the teaching process for more than three years, it can be concluded that the program effectively improves the learning effect of students, improves the teaching ideas and methods of teachers, and improves the level of teaching quality. It is worth popularizing widely in the teaching of discrete mathematics of computer major in Colleges and universities.

Conventional methods such as questionnaires and scales to evaluate learners' learning immersion are influenced by individuals' subjective factors. The non-synchronism between the learning state and after-learning investigation also reduces the accuracy. We propose a new method to evaluate learners' learning immersion based on electroencephalogram (EEG) and support vector machine (SVM). We construct 2 learning scenarios to induce immersive senses: VR video learning for high-level immersion and online English word learning for low-level immersion. To distinguish two immersion levels, students' EEGs are collected. After entering their attention score, relaxation score, the synchronization rate between the 2 scores, high alpha and low beta wave into SVM model, the precision accuracy reaches 87.80%. Taken the classified results and the participants' self-reports together, we find VR devices can create a more immersive environment which improves learners' learning effect. Our findings provide evidence supporting the feasibility of predicting learning immersion levels by physiological recordings.

It is an important aspect of the development of online education to record and analyze the educational data generated by online learners, predict the learning effect and provide precise and personalized services for them. In order to realize the prediction learning effect, firstly, the valuable information in the education data is mined and analyzed, and the data modeling is carried out. The data items are defined, and the solution is proposed according to the prediction results. Secondly, K-NN, LVQ and SVM are used to predict the learning effect. The experimental results show that the prediction accuracy of the three algorithms is as high as 70%. Finally, the three models are compared and analyzed, and learning behavior data are combined to provide a decision-making basis for the effective prediction of learning state, which is helpful for the online course platform to better plan students' learning plans and give corresponding learning warnings.

The attention flow network is an important branch of network science, and the research on the link prediction algorithm of attention flow network has important theoretical and application value. The traditional link prediction algorithm based on the neighbor hypothesis cannot learn the global structure information of the network, while the traditional graph neural network algorithm needs to combine external information and has a large amount of calculation. Based on the structural characteristics and the advantages of the MLP neural network, this paper proposes a stru2vec-based algorithm. The link prediction SMLP algorithm is compared with the existing DeepWalk, Line, Node2vec, Hope, NetMF and other algorithms. It is found that the SMLP algorithm proposed in this paper improves on the four evaluation indicators of Precision, Recall, Accuracy and F1 by about 15.8% on average. The average increase in AUC index is about 12.1%, and the performance is better on the ROC curve, and the algorithm proposed in this paper focuses on learning the network structure, and has a good learning effect for the network with missing node label information and sparse self-characteristic attributes.

This research examines senior high school students’ earth science learning effects, focusing on the influence of science reading beliefs when employing mobile devices. The revision of the Science Reader Belief Inventory (SRBI) was used to examine the connections for high school students’ personal scientific reading beliefs and reading comprehension of earth science learning effectiveness conditions when using mobile devices to learn. The learning outcome was determined using achievement tests. In this research, 97 students from three classes of first-year high school students were enrolled in an eight-week experimental teaching study followed by an achievement test. The major findings are as follows: (1) High school students’ transaction beliefs were stronger than transmission beliefs. Transaction beliefs were significantly correlated with transmission beliefs. (2) Scientific beliefs may take a long time to change. (3) Whereas traditional reading comprehension strategies seem to have relied more heavily on vocabulary development, in an e-learning environment, students tend to rely on sentence-level parsing to understand scientific texts. This research provides a reference for teachers within learning environments in which information is incorporated into technology instruction, and various learning scenarios are used.

This study analyzes the advantages and functions of mobile learning in college teaching under the hybrid teaching mode of adding mobile learning to conventional teaching. Examples show that mobile learning can make students more enthusiastic, more active and better learning effect. At the same time, it lists the shortcomings of mobile learning and gives the corresponding countermeasures.

Cryptography is the core of information security. Cryptography course is the central course in the construction of information security major. However, cryptography has a lot of theoretical knowledge, making students feel abstract and boring, and there is no suitable practical application, which leads to students' unsatisfactory learning effects. As the current advanced technology, blockchain is the perfect combination of cryptography and digital technology, but it is not integrated into the cryptography teaching. Therefore, in order to strengthen the training of information security talents, improve the effect of cryptography, and enhance students' interest in learning cryptography, this paper proposes a combination of blockchain technology for cryptography reform by experiential teaching, and the knowledge points in cryptography courses are targeted in learning in the blockchain. Taking the hash function in cryptography as an example, the “mining” of blockchain is experienced, and the important properties of the hash function, such as the antigenic image (unidirectional) and anti-collision, can deeply understood. At last, I hope to provide reference and new ideas for the construction of information security majors and training talent.

Federated learning enables participants to be capable of collaboratively building powerful machine learning models and exploiting privacy protection mechanisms to protect data privacy. However, the federal-learning security mechanism remains not perfect. Using malicious training samples is considered to be vulnerable to attacks against machine learning models. Lastly, the substitution technique of the model is adopted to eliminate the learning effect of the final model. Though numerous security protocols have been formulated to defend against and prevent malicious attacks, most are active defenses against malicious attacks, instead of passive ones (e.g., distillation-based defense and regularization-based adversarial training). In the present study, a block chain-based passive defense model is proposed to protect the model. To be specific, when a participant updates its local model, model participation, user fingerprints and other key information will be stored on the chain. Malicious attacks can be therefore traced and detected. To ensure data integrity and confidentiality, model data is encrypted and updated via the blockchain. Record rewards for active participation in federal learning and punish violators. Accordingly, the dynamic protection mechanism of federated learning is achieved.