International Journal of Continuing Engineering Education and Life-Long Learning (36 papers in press)

Regular Issues

• Teaching Practice-oriented Computer Vision Courses in COVID-19 Pandemic  
  by Jing Tian 
  Abstract: This paper aims to present an online teaching pedagogic experience for the practice-oriented computer vision course during the COVID-19 pandemic. COVID-19 has been disruptive to the education system worldwide, particularly to the computer vision course that usually requires face-to-face lectures and project collaboration during the study. This paper addresses three fundamental questions in teaching computer vision courses: 1) how to design the course topic and adapt to the online teaching format?; 2) how to conduct hybrid project collaboration in a hybrid mode?; 3) how to conduct the course assessment efficiently online? More specifically, this paper presents the pedagogic experience, including learning objectives, course curriculum structure, teaching methodologies, as well as final holistic assessments. The presented approach is an effective way of teaching practical computer vision courses, as verified by feedback from students. These experiences can be insightful to other lecturers who need to design, develop and deliver similar courses in the post pandemic era.
  Keywords: engineering education; computer vision teaching; active learning.
  DOI: 10.1504/IJCEELL.2022.10041479
   
  
• Research on College Students' Emotional Experience in Online Learning  
  by Lifeng Wang, Zi Ye, Shaotong Zhu 
  Abstract: Online education is the main form of pedagogy during the COVID-19 pandemic. The description and analysis of the emotional experience of online education for college students are of considerable significance. This paper surveyed and investigated the online learning experience of 1,147 college students during the pandemic. Moreover, their experience of offline and online learning was analysed. The results demonstrate that college students prefer face-to-face learning. Online learning produces more negative emotions, whereas face-to-face learning induces more positive emotions; academic situation, self-management, learning atmosphere, interpersonal relationship, self-expression, collective honour are the leading causes of students’ emotions.
  Keywords: face-to-face learning; online learning; emotional experience.
  DOI: 10.1504/IJCEELL.2023.10044197
   
  
• EXPLORATION OF GENDER SPECIFIC AND LEARNING ENVIRONMENT SPECIFIC COMFORT LEVEL IN COLLABORATIVE LEARNING  
  by Anitha D. Dhakshina Moorthy, Purnima Ahirao, Ramaa AnanthaMurthy, Vrinda Ullas 
  Abstract: Collaborative learning creates interest and engagement in learning. The collaborative learning activities can be conducted in one of the three learning environments or modes: face-to-face, completely online and blended mode and with different gender compositions. It is essential to understand the comfort level of students when working in different gender compositions and different learning environments. Hence, this research study is carried out with different group compositions in three different learning environments of collaborative learning: online, blended and face-to-face. One hundred seventy students from four educational institutions in India participated in this study. A survey is administered to them on the various parameters of comfortable learning. From the results obtained, it is observed that online or blended learning results in an increased comfort level of students than face-to-face learning sessions and the gender composition affects the comfort level.
  Keywords: collaborative learning; learning environments; gender; comfort; team formation.
  DOI: 10.1504/IJCEELL.2023.10044388
   
  
• The method of online classroom teaching quality evaluation based on deep data mining  
  by Jing Hou  
  Abstract: In order to overcome the problems of traditional online classroom teaching quality evaluation methods, such as low accuracy of quality evaluation and poor effect of classroom teaching quality improvement, this paper proposes an online classroom teaching quality evaluation method based on deep data mining. Fuzzy comprehensive evaluation method is used to quantify the evaluation index of online classroom teaching quality; The evaluation matrix is constructed to calculate the weight of classroom teaching quality evaluation index; The online classroom teaching quality evaluation indicators are classified by naive Bayes classification algorithm; With the help of deep data mining algorithm, this paper evaluates the post classification evaluation index, constructs the online classroom teaching quality evaluation model, and completes the online classroom teaching quality evaluation. The experimental results show that the accuracy of the proposed method is about 0.9, and it can effectively improve the quality of online classroom teaching.
  Keywords: Deep data mining; Classifier; Naive Bayes classification algorithm; Evaluation matrix.
  DOI: 10.1504/IJCEELL.2024.10046211
   
  
• Corpus-driven recommendation algorithm for English online autonomous learning resources  
  by Chao Han 
  Abstract: In order to overcome the problems of the traditional recommendation algorithm of English online learning resources, such as low accuracy, poor convergence and low success rate of recommendation. This paper proposes a recommendation algorithm for online language autonomous learning resources driven by CORUS. Based on the learning vector quantisation (LQN) algorithm, the model of subject word generation based on vector corpus resources is established. In the parameter training of the classification model, the vector weight is normalised to complete the optimisation of the corpus resource classification LVQ subject model. According to the binary particle swarm optimisation algorithm, the personalised recommendation model of autonomous learning resources is implemented. Experimental results show that the proposed vector quantisation network algorithm has high convergence, and the recommended success rate is 99.5%. Therefore, the method proposed in this paper can effectively complete the recommendation of English online autonomous learning resources.
  Keywords: corpus; English learning; learning resources; recommendation algorithm; binary particle swarm.
  DOI: 10.1504/IJCEELL.2024.10046212
   
  
• Effect of personal response systems on students' academic performance and perception in online teaching  
  by Hongjiang Wang, Zuokun Li, Xiaojin Wang 
  Abstract: To minimise the negative impact of the epidemic on education, online teaching was taking off all over the world. In order to improve the interactive effect of online teaching, some scholars have applied personal response systems (PRSs) during the online teaching process. However, there is few experimental studies to validate the effectiveness of PRSs, such as students performance and perceptions. In the paper, based on the quantitative and qualitative mixed analysis method, which is the best way to gain information of specific experience, 98 sophomores majoring in Applied Mathematics are sampled to test students academic performance and perception in online teaching using PRSs, e.g., UMU. Some conclusions can be drawn: 1) the experimental groups academic performance was higher than the control group; however, there was no significant difference between them; 2) majority of students have positive perceptions, such as learning motivation (LM), learning engagement (LE) and learning satisfaction (LS) towards PRSs. And there was no significant difference in gender perception. At last, the paper concluded the advantages, disadvantages and suggestions of PRSs used in online teaching. The limitation of research and the direction of future research are discussed.
  Keywords: personal response systems; PRSs; academic performance; perception; online teaching.
  DOI: 10.1504/IJCEELL.2023.10046526
   
  
• Investigating the Effect of Online Teaching using a SPOC  
  by Hongtao Yu 
  Abstract: The outbreak of COVID-19 in 2019 brought about unprecedented changes to higher education. Following the deployment of the Education Ministry, Chinese higher learning institutions carried out practical activities in response to the classes suspended but learning continues appeal by making use of internet technologies. In order to understand the effect of online teaching, this research uses qualitative research methods to study online learning of students in a certain university, and selects four case classes with different learning abilities. After conducting a comparative analysis of final exam scores, student questionnaire surveys and interviews, analysis of the effectiveness of students online learning of the course, the results show that students online learning performance is positively correlated with the scores of the preceding courses, and the online learning duration is positively correlated. Online learning is more suitable for students with strong self-discipline and three suggestions are given for effective teaching in the future.
  Keywords: SPOC; online learning; empirical research.
  DOI: 10.1504/IJCEELL.2023.10046527
   
  
• Multi-dimensional Dynamic Evaluation of MOOC English Mixed Teaching Based on BP Neural Network  
  by Mian Wang 
  Abstract: In order to overcome the problem of low accuracy in current evaluation methods of MOOC English mixed teaching, this paper proposes a multi-dimensional dynamic evaluation method based on BP neural network. By collecting evaluation data from teaching experts, teachers and students, the basic dataset of English teaching evaluation is constructed. The data from the evaluation basic dataset were taken as input samples, and the input samples were normalised. The input samples were input into the constructed BP neural network evaluation model, and the multi-dimensional dynamic evaluation results of MOOC mixed English teaching were output. Experimental results show that the evaluation accuracy of the proposed method is more than 90%, and the convergence can be achieved only for about 50 times, the convergence speed is faster and the evaluation time is shorter.
  Keywords: BP neural network; MOOC English teaching; multi-dimensional dynamic evaluation; Basic dataset.
  DOI: 10.1504/IJCEELL.2024.10047879
   
  
• Sustaining College Students’ Continuance Intention Toward Online Learning in the Post-COVID-19 Era  
  by Minghua He, Jialiang Qin, Rongfang Tang 
  Abstract: Since the outbreak of the COVID-19 epidemic, universities in China have integrated online teaching, and a total of 17.75 million college students across the country have participated in these virtual learning systems. The extensive growth in online education that we have witnessed is unprecedented in the history of higher education. Based on the technology acceptance model (TAM), this study investigates the impacts of perceived enjoyment, ease of use, and usefulness on Chinese college students’ continuance intention toward online learning. Our empirical findings indicated that the continuance intention of college students was affected by these perception factors and that innovativeness played a moderating role in the relationship between perceived enjoyment and continuance intention and perceived usefulness and continuance intention. These results provide theoretical and practical insights for universities and online education platforms that can be used to stimulate endogenous student motivation and sustain online learning behaviour in the post-COVID-19 era.
  Keywords: online learning; online education; college students; continuance intention; innovativeness; perceived enjoyment; COVID-19; technology acceptance model; TAM.
  DOI: 10.1504/IJCEELL.2024.10048378
   
  
• MOOC English online course recommendation algorithm based on LDA user interest model  
  by Zhongping Yao 
  Abstract: In order to improve the efficiency and accuracy of course recommendation and improve user satisfaction, a MOOC English online course recommendation algorithm based on LDA user interest model is proposed. Wavelet transform method is used for data denoising to improve the accuracy of recommendation results; Using support vector machine to classify courses to improve the efficiency of course recommendation; LDA user interest model is established to describe the characteristics of students’ online learning behaviour. According to the characteristics of students’ interest and learning behaviour, the matching topics can be selected to realise English online course recommendation. The experimental results show that the highest accuracy of course recommendation of this method is 92%, and the student satisfaction basically reaches more than 90 points, which verifies the effectiveness of this method.
  Keywords: LDA user interest model; course recommendation; wavelet transform; support vector machine; SVM; data denoising.
  DOI: 10.1504/IJCEELL.2024.10050126
   
  
• MOOC distance teaching effect evaluation method based on fuzzy entropy  
  by QingQin Chen 
  Abstract: In order to overcome the problems of low evaluation accuracy and long evaluation time in traditional evaluation methods, a MOOC distance teaching effect evaluation method based on fuzzy entropy is proposed. Firstly, mining MOOC distance learning data. Secondly, according to the needs of teaching effect evaluation, build the MOOC distance teaching effect evaluation index system. Finally, according to the principle of fuzzy entropy, the fuzzy entropy weight of the evaluation index is calculated, the fuzzy entropy weight is normalised, and the attribute matrix of the evaluation index is constructed. The ideal point and closeness degree are calculated according to the attribute matrix, and the effect of MOOC distance teaching is evaluated through the closeness degree. The experimental results show that compared with the traditional evaluation methods, this method greatly improves the evaluation accuracy on the basis of reducing the evaluation time, and the maximum evaluation accuracy is 97%.
  Keywords: fuzzy entropy; MOOC; distance learning; impact assessment.
  DOI: 10.1504/IJCEELL.2024.10050127
   
  
• An optimization of higher education resources search method based on multi-state hierarchical model  
  by Ping Li  
  Abstract: In order to overcome the problems of low recall rate, precision rate and large search time consumption of traditional methods, optimisation of higher education resources search method based on multi-state hierarchical model is studied. This paper analyses the objective function of higher education resources search, sets the related constraint conditions, and selects the higher education resources search mode by ant colony algorithm. In order to further improve the search quality, the ant colony algorithm was improved by selecting ant colony species, determining communication subgroups, communication period and information exchange between subgroups, and the algorithm was used to optimise the search mode, that is, resource stratified search. A multi-state hierarchical model is built to search higher education resources. Experimental results show that the recall rate of this method is always above 93%, the precision rate is above 92%, and the average search time consumption is 0.66 s.
  Keywords: multi-state hierarchical model; higher education resources; search method optimisation; ant colony algorithm; heterogeneous multiple ant colony algorithm.
  DOI: 10.1504/IJCEELL.2024.10050128
   
  
• University online teaching resource sharing open platform based on deep learning  
  by Liangxi Ding  
  Abstract: Aiming at the problems of high energy consumption, low classification accuracy of shared data and poor sharing effect of open network teaching resource sharing platform, an open network teaching resource sharing platform based on deep learning is designed. Setup application module, database module and database retrieval function module in the platform, classify online teaching resources in colleges and universities by using deep learning algorithm, determine the characteristics of online teaching resources in colleges and universities, and build an open platform for sharing online teaching resources in colleges and universities. The experimental results show that the platform designed in this paper has low energy consumption, which is always lower than 20j, and the data classification accuracy of shared online teaching resources is always higher than 90%, which can effectively improve the sharing effect of online teaching resources in colleges and universities, and has good practical application performance.
  Keywords: deep learning; application module; database module; database retrieval function module; softmax regression.
  DOI: 10.1504/IJCEELL.2024.10051304
   
  
• A comprehensive evaluation of network teaching quality in colleges and universities based on entropy weight TOPSIS model  
  by QingQin Chen 
  Abstract: Because the traditional evaluation methods have the problems of low weight calculation accuracy and high evaluation error rate, a comprehensive evaluation method of network teaching quality in colleges and universities based on entropy weight TOPSIS model is proposed. Firstly, questionnaire survey, expert interview and statistical methods are used to obtain the data related to network teaching. Secondly, the evaluation index system is established according to relevant principles. Finally, the entropy weight method is used to determine the weight of the evaluation index, combined with the calculation results of the weight of index, the TOPSIS model is used to get the relevant evaluation results. The experimental results show that this method can realise the evaluation of network teaching quality in colleges and universities. The average accuracy of evaluation index weight calculation is 95.9%, and the average evaluation error rate is 7.6%.
  Keywords: entropy weight method; TOPSIS model; teaching quality evaluation; questionnaire survey; expert interview; statistical method.
  DOI: 10.1504/IJCEELL.2024.10051305
   
  
• An Online learning behavior monitoring of students based on face recognition and feature extraction  
  by Dong-yuan Ge, Jian Li, Hai-ping Luo, Tuo Zhou, Wen-jiang Xiang, Xi-fan Yao 
  Abstract: In order to effectively improve the accuracy and efficiency of students’ online learning behaviour monitoring, an online learning behaviour monitoring method based on face recognition and feature extraction is proposed. Analyse the relevant theories of face recognition technology and feature extraction methods, and collect the global features of students’ online learning behaviour by monitoring video images and using face recognition technology. Using the feature extraction method, the local features of students’ online learning behaviour are extracted according to the grey value of video pixels. On this basis, it constructs the monitoring model of students’ online learning behaviour to realise the monitoring of students’ online learning behaviour. The experimental results show that the proposed method has good monitoring effect on students’ online learning behaviour, and can effectively improve the accuracy and efficiency of students’ online learning behaviour monitoring. The maximum monitoring accuracy of the proposed method is more than 97%.
  Keywords: face recognition technology; feature extraction method; online learning; behaviour monitoring.
  DOI: 10.1504/IJCEELL.2024.10051306
   
  
• An English teaching resource database retrieval based on adaptive differential evolution algorithm  
  by Mengzhang Liu 
  Abstract: In order to improve the retrieval recall rate of English teaching resource database, improve the retrieval accuracy of English teaching resource database and shorten the retrieval time of English teaching resource database, the English teaching resource database retrieval method based on adaptive differential evolution algorithm is studied. Firstly, the semantic feature distribution of English teaching resource database is analysed by combining semantic information fusion and cluster analysis. Then, the spatial undersampling technology is used to process the distributed sampling and information fusion of the English teaching resource database. Finally, according to the fusion results, the adaptive differential evolution algorithm is used to retrieve the English teaching resource database. The experimental results show that the average recall rate of the proposed method is 90.7%, the retrieval accuracy of English teaching resource database is 97.7%, and the retrieval time is 22.7 ms.
  Keywords: adaptive differential evolution algorithm; cluster analysis; semantic information fusion; English teaching; resource database retrieval.
  DOI: 10.1504/IJCEELL.2024.10051372
   
  
• Study on the differences in behaviour characteristics of distance online autonomous learning in different network virtual environments  
  by Huan Wu 
  Abstract: Due to the complexity of e-learning mode, the behaviour characteristics of online autonomous learning are different, and the existing research methods cannot comprehensively study the behaviour characteristics. Therefore, this paper puts forward the research on the behaviour characteristics of distance online autonomous learning in different network virtual environments. This paper uses data mining technology to mine the data of online autonomous learning behaviour characteristics, analyses the general characteristics of different types of students’ learning behaviour, summarises the influencing factors and characteristic dimensions of online autonomous learning mode, and analyses the influence of learners’ personality characteristics on distance online autonomous learning. Through the questionnaire survey experiment, the reliability coefficient of this method in analysing the differences of autonomous learning behaviour characteristics is calculated. The reliability coefficient of this study is above 0.786, which can better grasp online learning resources and improve students’ online autonomous learning ability.
  Keywords: network virtual environment; autonomous learning; behaviour characteristics; behaviour differences.
  DOI: 10.1504/IJCEELL.2024.10051682
   
  
• MOOC English Online Learning Resource Recommendation Algorithm Based on spectral clustering and matrix decomposition  
  by Qichao Huang  
  Abstract: In order to overcome the problems of low recommendation accuracy and long recommendation time in traditional MOOC English online learning resource recommendation algorithm, a new MOOC English online learning resource recommendation algorithm based on spectral clustering and matrix decomposition is proposed. The clustering objective function is constructed by spectral clustering method to complete the clustering of MOOC English online learning resources. Based on the results of resource clustering, the objective matrix, row auxiliary matrix and column auxiliary matrix are constructed. The matrix decomposition method is used to construct the recommended scoring matrix, and the scoring matrix is filled and reduced to complete the recommendation of MOOC English online learning resources. Experimental results show that, compared with the traditional learning resource recommendation, the proposed algorithm has higher clustering accuracy, higher recommendation accuracy and efficiency, and the maximum recommendation time is only 1.1 s.
  Keywords: spectral clustering; matrix decomposition; MOOC English online learning; learning resources recommendation.
  DOI: 10.1504/IJCEELL.2024.10052668
   
  
• An online teaching quality evaluation method based on deep belief network  
  by Chun LIANG, Hai-lin PENG 
  Abstract: In order to improve the evaluation effect and accuracy of online teaching quality, an online teaching quality evaluation method based on deep belief network is proposed. Establish the evaluation index system of online teaching quality, collect the data related to teaching quality, teaching attitude, teaching content, teaching methods and teaching influence by using crawler technology, and extract the data characteristics of online teaching quality evaluation index. Combined with the data characteristics, the online teaching quality evaluation model is constructed by using the deep belief network, and the evaluation index data is input into the evaluation model to obtain the online teaching quality score. The experimental results show that the error rate of the proposed method is only 4.9%, the average accuracy rate of online teaching quality evaluation is as high as 97.2%, which has the characteristics that the accuracy of online teaching quality evaluation is higher than the effect.
  Keywords: online teaching; deep belief network; DBN; teaching quality evaluation; restricted Boltzmann machine; RBM; evaluation index system.
  DOI: 10.1504/IJCEELL.2024.10052683
   
  
• Study on quality evaluation of online and offline mixed teaching reform based on big data mining  
  by Guoxia Hu, Suntai Sun, Zhongxiao Sun 
  Abstract: In order to improve the accuracy of the reform quality research and shorten the overall research time, the reform quality research is carried out based on the big data mining technology. First, the local density information of the data is calculated and the required samples are mined. Secondly, the probabilistic undirected graph model is used to remove the noise in the mining samples and improve the accuracy of the sample data. Finally, the PCA algorithm in big data is used to calculate the contribution rate of the sample data, and the reform evaluation model is constructed. The test results of different indicators show that the accuracy rate of the research method is 92.6%, and the evaluation time is only 12.7 s, which can effectively improve the evaluation accuracy and shorten the evaluation time.
  Keywords: big data mining; online and offline mixed teaching; PCA algorithm; reform in education; quality assessment.
  DOI: 10.1504/IJCEELL.2024.10053206
   
  
• Online and offline hybrid teaching data mining based on decision tree classification  
  by Yu Cao, Shu-wen Chen, Hui-sheng Zhu 
  Abstract: In order to overcome the problems of large mining errors and low classification accuracy of traditional teaching data mining methods, a hybrid online and offline teaching data mining method based on decision tree classification is proposed. First of all, the online and offline mixed teaching data is obtained with the help of crawler technology. Secondly, data repair method is adopted to ensure data consistency, and duplicate data values are determined by distance value to complete data pre-processing. Finally, according to the construction of the decision tree, determine the root entropy and leaf entropy of the mixed teaching data, create the root node, attribute list and class list of the mixed teaching data, and complete the online and offline mixed teaching data mining. The experimental results show that the proposed method can effectively reduce the error of data mining, with the error coefficient not exceeding 0.2, and improve the classification accuracy.
  Keywords: decision tree classification; online and offline teaching; data mining: crawler technology; root entropy; leaf entropy.
  DOI: 10.1504/IJCEELL.2024.10053207
   
  
• An Evaluation method of online education reform effect based on fuzzy weight  
  by Hui Lv, Mingyang Gao, Tian Hong 
  Abstract: In order to overcome the problems of large calculation error and low evaluation accuracy of traditional online education reform effect evaluation methods, this paper proposes a new online education reform effect evaluation method based on fuzzy weight. First, the key influencing factors of online teaching reform effect evaluation are determined for different subjects. Then, the cluster algorithm is used to determine the cluster centre of the evaluation index data, and the construction and quantification of the online teaching reform effect evaluation index system are completed. Finally, the fuzzy weight is determined, and the online education reform effect evaluation algorithm is constructed by using the judgment matrix and the training evaluation index data set. The experimental results show that this method can reduce the calculation error of evaluation weight and improve the evaluation accuracy, and the evaluation accuracy is always kept above 90%.
  Keywords: fuzzy weight; online education reform; effect evaluation; clustering algorithm; product quantification model; judgement matrix.
  DOI: 10.1504/IJCEELL.2024.10053208
   
  
• Multi dimensional dynamic evaluation method of English MOOCS autonomous learning based on Multiple Intelligences Theory  
  by Lizhen Shi 
  Abstract: In order to overcome the problems of traditional English autonomous learning evaluation methods, such as poor multi-dimensional evaluation effect, long evaluation time and inaccurate evaluation results, this paper proposes a multi-dimensional dynamic evaluation method of English autonomous learning based on the theory of multiple intelligences. The optimal multi-dimensional evaluation solution of English MOOCS autonomous learning is obtained by the theory of multiple intelligences, the optimal classification surface of English MOOCS autonomous learning is obtained by Lagrange function, and the multi-dimensional dynamic evaluation matrix is obtained by binary tree; This paper uses Euclidean distance binary tree support vector machine (DBT-SVM) multi class classification algorithm to evaluate English MOOCS learning. The experimental results show that the multi-dimensional dynamic evaluation of autonomous learning takes the least time, only 4.2 s, and the accuracy rate of multi-dimensional dynamic evaluation is as high as 97%.
  Keywords: binary tree; theory of multiple intelligences; Lagrange function; multidimensional dynamic evaluation matrix.
  DOI: 10.1504/IJCEELL.2024.10053209
   
  
• Study on Accurate Evaluation of Network Distance Education Quality Based on Analytic Hierarchy Process  
  by You Yin 
  Abstract: In order to overcome the problems of low evaluation accuracy and low fitting degree of evaluation model existing in traditional methods, a new accurate evaluation method of network distance education quality based on analytic hierarchy process was proposed. Data mining method is used to obtain quality evaluation data of network distance education, and fuzzy neural network is used to normalise the mined data. The evaluation system of education quality was established, and all indicators were compared pairwise by AHP. Meanwhile, the comparison results were transformed into a comparative judgement matrix, and the calculation results of index weights were obtained, so as to construct an accurate evaluation model of education quality. The experimental results show that the evaluation accuracy of the proposed method is high, and the fitting accuracy of the evaluation model is more than 95%, and the fitting time is always kept below 15 s.
  Keywords: analytic hierarchy process; network distance education; quality of education; accurate evaluation; fuzzy neural network.
  DOI: 10.1504/IJCEELL.2024.10053210
   
  
• Deep Mining of Mobile Learning Data Based on Multi-scale Clustering Analysis  
  by Yuxin Ji 
  Abstract: In order to overcome the problems of low accuracy and long time of traditional mobile learning data mining methods, this paper proposes a new mobile learning data deep mining method based on multi-scale clustering analysis. The original dataset was constructed according to the database of the mobile learning platform, and the data in the original dataset was merged and the missing value was filled. On this basis, the data processing results are divided into multiple scales, and the appropriate benchmark scale is selected for scale clustering, so as to carry out the depth mining of mobile learning data. Experimental results show that the accuracy of mobile learning data mining of the proposed method is always above 95%, the average mining time is only 0.59 s, and the stability of the mining process is good, so it can be applied and promoted in practice.
  Keywords: multi-scale clustering analysis; mobile learning; data depth mining; missing value filling.
  DOI: 10.1504/IJCEELL.2024.10053211
   
  
• Research on English Teaching Reading Quality Assessment Based on Cognitive Diagnostic Assessment  
  by Xiangying KOU, Zhenhong Yang, Yuyu Wang 
  Abstract: In order to overcome the problem that the traditional English teaching reading quality evaluation method does not pay attention to the influence of students’ psychological cognition on the evaluation, which leads to the poor practical application effect, a research on the evaluation method of English teaching reading quality based on cognitive diagnostic evaluation is proposed. This method constructs English reading cognitive attributes, introduces Q-matrix and answers data, constructs English reading cognitive diagnosis model, uses fuzzy comprehensive evaluation method, divides evaluation index levels, and generates diagnosis feedback results. Under the guidance of cognitive diagnosis theory, derive knowledge structure and cognitive strategies. The experimental results show that the sampling results of this method are reasonable, and the difficulty level relationship of mental representations is relatively consistent, which effectively improves the quality of English teaching reading, and provides strong evidence for the validity of the cognitive model of reading ability.
  Keywords: cognitive diagnosis assessment; English teaching; reading quality; assessment.
  DOI: 10.1504/IJCEELL.2024.10053212
   
  
• Collaborative recommendation model of MOOC online learning resources based on scoring matrix  
  by Jun Yao 
  Abstract: In this paper, a MOOC online learning resource collaborative recommendation model based on scoring matrix is proposed. The features of online learning resources and learners’ learning level are determined, and the differences between them are processed by kernel function to extract the features of online learning resources; The data recommendation matrix is constructed through the score matrix, and the missing characteristic data values in the score matrix are expressed. The data weight of online learning resources is calculated, and the MOOC online learning resources collaborative recommendation model is designed. The experimental results show that the accuracy of MOOC online learning resources recommended by the proposed model is always higher than 90%.
  Keywords: online learning resources; collaborative recommendation model; Jaccard coefficient; scoring matrix.
  DOI: 10.1504/IJCEELL.2024.10053213
   
  
• Deep mining method of online learning behavior data based on big data analysis  
  by Weijuan Li 
  Abstract: Aiming at the problems of low mining accuracy and long mining time in learning behaviour data mining, a deep mining method of online learning behaviour data based on big data analysis is proposed. The initial signals of learners’ subject, object, learning environment, learning means, time and result data are set, and the data components of online learning behaviour are obtained through EMD; EEMD is used to extract the key features of online learning behaviour data, and different contribution rates in learning sequence are calculated by linear weighting method; With the help of the first-order polynomial decision function in big data technology, the inner product is calculated, and the online learning behaviour data deep mining model based on big data technology is constructed to complete the data deep mining. The experimental results show that the accuracy of the proposed method is about 95%, and the mining time is short.
  Keywords: big data analysis; online learning behaviour data; depth mining; EEMD; linear weighting; depth mining model.
  DOI: 10.1504/IJCEELL.2024.10053214
   
  
• Big data classification of learning behavior based on data reduction and ensemble learning  
  by Taotao Wang, Xiaoxuan Wu 
  Abstract: In order to overcome the problems of low classification accuracy, long time and high missing ratio of traditional methods, a big data classification method of learning behaviour based on data reduction and ensemble learning was proposed. By cleaning and transforming the big data of learning behaviour and discretising the attributes of big data of learning behaviour, the data reduction algorithm is used to simplify the attributes of big data of learning behaviour. The ensemble learning method is used to linearly combine several weak classifiers, and the ensemble classifier is trained according to Choquet integral. The trained classifier is used to classify the big data of learning behaviour after simplified processing. The experimental results show that when the amount of big data on learning behaviour reaches 5,000 GB, the average classification accuracy of the proposed method is 92%, the classification time is 29 s, and the failure rate of classification is 0.32%.
  Keywords: data reduction; ensemble learning; rough set theory; big data of learning behaviour; big data classification.
  DOI: 10.1504/IJCEELL.2024.10053215
   
  
• Evaluation method of online education quality based on Fuzzy Rough Set  
  by Ying Zhou, Lei Zhang 
  Abstract: In order to overcome the problems of unreasonable index weight distribution, poor quality evaluation results and low accuracy in the process of online education quality evaluation, this paper proposes a method of online education quality evaluation based on fuzzy rough set. Fuzzy rough set theory is introduced to eliminate redundant attributes of network. According to the regional characteristics of online education network, this paper constructs an evaluation index system, calculates the evaluation weight of education network, evaluates the importance of education network nodes, determines the level of education quality, and obtains the evaluation results of education quality. The experimental results show that the maximum value of evaluation index weight distribution rationality parameter can reach 29.45, and the quality evaluation result U≥90, which shows that the quality effect is excellent, and the accuracy rate reaches 90%. This method has good practicability and guidance.
  Keywords: fuzzy rough set; online education; quality evaluation.
  DOI: 10.1504/IJCEELL.2024.10053216
   
  
• Evaluation Method of Immersive Situation Interpretation Teaching Based on Natural Language Processing
  by Weihua Wang 
  Abstract: In order to overcome the problems of poor evaluation accuracy and low evaluation efficiency, this paper proposes an immersive situational interpretation teaching effect evaluation method based on natural language processing. Firstly, the evaluation system of interpretation teaching should be established; secondly, it constructs a set of factors for interpreting evaluation and determines the evaluation criteria for interpreting teaching difficulty; then, according to the natural language processing method, calculate the similarity of the evaluation standards at all levels; finally, a teaching effect evaluation function is constructed to evaluate the teaching effect of immersive situational interpretation. The experimental results show that the evaluation results of this method are closer to the students’ scores, the accuracy of the effect evaluation is improved by 8%, and the evaluation time is shortened by about 5 s, indicating that the accuracy of the evaluation results of this method is higher.
  Keywords: natural language processing; situational interpretation; teaching effect evaluation; objective weighting method; automatic word segmentation.
  
  
• Data mining method of English autonomous learning behavior based on decision tree
  by Juan Zhang, Pingyang Li, Xiaoli Sun 
  Abstract: Because the traditional data mining methods of English autonomous learning behaviour have the problems of low accuracy and long mining time, a decision tree-based data mining method of English autonomous learning behaviour is proposed. Firstly, the students’ learning behaviour data is collected, and then the collected behaviour data is classified by the decision tree method, and the data is divided into different types. Finally, according to the data classification results, the cart decision tree method is used to obtain the optimal split point of the decision tree through tree building and pruning operations, and the optimal binary tree is generated to realise the data mining of English autonomous learning behaviour. The experimental results show that the highest comprehensive coefficient of data mining of the research method in this paper is increased by 0.08 and 0.04 respectively, and the accuracy and efficiency of data mining are improved.
  Keywords: decision tree; learning behaviour; data mining; data acquisition platform; information gain; collaborative filtering; cart decision tree; prune.
  
  
• Research on quality evaluation of teaching reform based on Cauchy function
  by Dakai Li, Liu Yang 
  Abstract: In order to improve the recall of the evaluation results and the accuracy of the feature clustering of the teaching reform data, this paper designs a teaching reform quality evaluation method based on Cauchy function. Firstly, mining the characteristics of teaching reform evaluation data, and using TOPSIS analysis method to score the evaluation indicators, establish the evaluation indicator system. Secondly, a judgment matrix is constructed for the evaluation index system, and the weight of the index is determined according to the importance of the evaluation index. Finally, taking Cauchy function as membership function, the final evaluation result is obtained through fuzzy integration. The experimental results show that with the increase of the number of experiments, the precision of the evaluation results obtained by this method is always above 95%, and the maximum clustering accuracy of the teaching reform feature data can reach 97%.
  Keywords: Cauchy function; teaching reform; quality assessment; TOPSIS analysis method; index weight; membership function.
  
  
• Evaluation algorithm of online and offline mixed teaching quality based on multivariate statistical analysis
  by Shijuan Shen, Qingqing Shi, Xiaojing Bai 
  Abstract: In order to improve the accuracy of teaching quality evaluation and reduce the evaluation time, an online and offline mixed teaching quality evaluation algorithm based on multivariate statistical analysis is designed. Association rules are used to set the influencing factors to determine the rule conditions, and to determine the influencing factors of teaching quality. The weight of influencing factors is calculated, and Lagrange multiplier is introduced to determine the influencing factors. The influencing factors are grouped by factor analysis method, and the determined influencing factors with high correlation are classified by cluster analysis. The discriminant function criterion is constructed by discriminant analysis, and the discrimination and evaluation of different influencing factors are realised. The experimental results show that the highest evaluation accuracy of the method in this paper reaches 97%, indicating that it effectively improves the accuracy of the evaluation and reduces the evaluation time.
  Keywords: multivariate statistical analysis; online and offline mixed teaching; quality assessment; Lagrange multiplier; discriminant function.
  
  
• Cloud computing based method for optimal allocation of college network course education resources
  by HaiHua Huang, XinBin Yang 
  Abstract: To improve the classification accuracy of online course education resources and reduce the time consumption in the process of optimal allocation of resources, this paper proposes a method of optimal allocation of online course education resources in colleges and universities based on cloud computing. The cloud platform for the allocation of college online course education resources is built, and the data of college online course education resources are obtained by LDA topic function. The adaptation decision of college online course education resources is designed, and the optimal allocation of college online course education resources is realised according to the cloud computing method. The experimental results show that the proposed method takes less than 3.9 s to optimise the allocation of 1,200 G online course education resources, the classification accuracy can reach 99.0%, and the allocation efficiency is effectively improved, indicating that the application effect of this method is good.
  Keywords: Shannon formula; cloud computing; LDA topic function; allocation of educational resources; adaptation decisions.
  
  

Special Issue on: Technology and Innovation Management in Education

• Study on abnormal behavior recognition of MOOC online English learning based on multidimensional data mining  
  by Fengxiang Zhang, Feifei Wang 
  Abstract: In order to overcome the problems of low recognition accuracy and long recognition time of traditional English learning abnormal behaviour recognition methods, this paper proposes MOOC online English learning abnormal behaviour recognition method based on multidimensional data mining. Firstly, set up the multi-dimensional association item set of MOOC online English learning behaviour, mine the learning behaviour data for correction. Secondly, students’ MOOC online English learning behaviour characteristics are extracted from students’ target contour and blinking behaviour characteristics. Then, taking this as the training sample subset, the individual member classifier is constructed by the mixed perturbation method to classify the learning behaviour. Finally, the abnormal behaviour identification of MOOC online English learning is completed. The experimental results show that the proposed method has high accuracy and short recognition time.
  Keywords: multidimensional data mining; MOOC online English learning; abnormal behaviour; mixed perturbation method; individual member classifier.
  DOI: 10.1504/IJCEELL.2022.10047472