E-learning behavior data indicates several students’activities on the e-learning platform such as the number of accesses to a set of resources and number of participants in lectures.This article proposes a new analyt...E-learning behavior data indicates several students’activities on the e-learning platform such as the number of accesses to a set of resources and number of participants in lectures.This article proposes a new analytics systemto support academic evaluation for students via e-learning activities to overcome the challenges faced by traditional learning environments.The proposed e-learning analytics system includes a new deep forest model.It consists of multistage cascade random forests with minimal hyperparameters compared to traditional deep neural networks.The developed forest model can analyze each student’s activities during the use of an e-learning platform to give accurate expectations of the student’s performance before ending the semester and/or the final exam.Experiments have been conducted on the Open University Learning Analytics Dataset(OULAD)of 32,593 students.Our proposed deep model showed a competitive accuracy score of 98.0%compared to artificial intelligence-based models,such as ConvolutionalNeuralNetwork(CNN)and Long Short-TermMemory(LSTM)in previous studies.That allows academic advisors to support expected failed students significantly and improve their academic level at the right time.Consequently,the proposed analytics system can enhance the quality of educational services for students in an innovative e-learning framework.展开更多
With this work, we introduce a novel method for the unsupervised learning of conceptual hierarchies, or concept maps as they are sometimes called, which is aimed specifically for use with literary texts, as such disti...With this work, we introduce a novel method for the unsupervised learning of conceptual hierarchies, or concept maps as they are sometimes called, which is aimed specifically for use with literary texts, as such distinguishing itself from the majority of research literature on the topic which is primarily focused on building ontologies from a vast array of different types of data sources, both structured and unstructured, to support various forms of AI, in particular, the Semantic Web as envisioned by Tim Berners-Lee. We first elaborate on mutually informing disciplines of philosophy and computer science, or more specifically the relationship between metaphysics, epistemology, ontology, computing and AI, followed by a technically in-depth discussion of DEBRA, our dependency tree based concept hierarchy constructor, which as its name alludes to, constructs a conceptual map in the form of a directed graph which illustrates the concepts, their respective relations, and the implied ontological structure of the concepts as encoded in the text, decoded with standard Python NLP libraries such as spaCy and NLTK. With this work we hope to both augment the Knowledge Representation literature with opportunities for intellectual advancement in AI with more intuitive, less analytical, and well-known forms of knowledge representation from the cognitive science community, as well as open up new areas of research between Computer Science and the Humanities with respect to the application of the latest in NLP tools and techniques upon literature of cultural significance, shedding light on existing methods of computation with respect to documents in semantic space that effectively allows for, at the very least, the comparison and evolution of texts through time, using vector space math.展开更多
The increased ownership of mobile phone users and the advancement of mobile applications enlarge the practicality and popularity of use for learning purposes among Chinese university students.However,even if innovativ...The increased ownership of mobile phone users and the advancement of mobile applications enlarge the practicality and popularity of use for learning purposes among Chinese university students.However,even if innovative functions of these applications are increasingly reported in relevant research in the education field,little research has been in the application of spoken English language.This paper examined the effect of using a Mobile-Assisted Language Learning(MALL)application“IELTS Liulishuo”(speaking English fluently in the IELTS test)as a unit of analysis to improve the English-speaking production of university students in China.The measurement of this mobile application in its effectiveness of validity and reliability is through the use of seven dimensional criteria.Although some technical and pedagogical issues challenge adoptions of MALL in some less-developed regions in China,the study showed positive effects of using a MALL oral English assessment application characterised with Automatic Speech Recognition(ASR)system on the improvement of complexity,accuracy,and fluency of English learners in China’s colleges.展开更多
目的开发一种深度学习系统用于成人发育性髋关节发育不良(Developmental dysplasia of the hip,DDH)患者的Crowe分型辅助诊断,并且分析该系统对于帮助临床医学生掌握DDH分型的可行性。方法纳入149例X线片训练集、42例测试集以及21例验证...目的开发一种深度学习系统用于成人发育性髋关节发育不良(Developmental dysplasia of the hip,DDH)患者的Crowe分型辅助诊断,并且分析该系统对于帮助临床医学生掌握DDH分型的可行性。方法纳入149例X线片训练集、42例测试集以及21例验证集,分割盆骨、提取DDH局部图像块,将金标准结果与医学生、AI辅助医学生评估结果进行比较。结果测试集共纳入42例,其中女性30例,男性12例,年龄(69±12)岁,涉及发育不良髋关节67侧(左30侧,右37侧)。AI、医学生、AI辅助医学生评估结果与金标准的相关性为0.906[95%CI(0.850,0.941)]、0.823[95%CI(0.726,0.887)]、0.886[95%CI(0.821,0.929)];准确率分别为0.87、0.78、0.88;精确度分别为0.88、0.83、0.89;召回率分别为0.87、0.78、0.88;F1值分别为0.87、0.80、0.88。混淆矩阵和条件概率结果显示,预测准确率Ⅰ型DDH三组分别为0.98、0.88、0.96,Ⅱ型DDH三组分别为0.40、0.20、0.40,Ⅲ型DDH三组分别为0.56、0.67、0.78;Ⅳ型DDH三组分别为0.88、0.75、0.88。结论深度学习辅助诊断系统可以有效提高医学生对于DDH分型的评估能力,可作为医学生学习掌握DDH影像诊断的培训工具。展开更多
基金The authors thank to the deanship of scientific research at Shaqra University for funding this research work through the Project Number(SU-ANN-2023017).
文摘E-learning behavior data indicates several students’activities on the e-learning platform such as the number of accesses to a set of resources and number of participants in lectures.This article proposes a new analytics systemto support academic evaluation for students via e-learning activities to overcome the challenges faced by traditional learning environments.The proposed e-learning analytics system includes a new deep forest model.It consists of multistage cascade random forests with minimal hyperparameters compared to traditional deep neural networks.The developed forest model can analyze each student’s activities during the use of an e-learning platform to give accurate expectations of the student’s performance before ending the semester and/or the final exam.Experiments have been conducted on the Open University Learning Analytics Dataset(OULAD)of 32,593 students.Our proposed deep model showed a competitive accuracy score of 98.0%compared to artificial intelligence-based models,such as ConvolutionalNeuralNetwork(CNN)and Long Short-TermMemory(LSTM)in previous studies.That allows academic advisors to support expected failed students significantly and improve their academic level at the right time.Consequently,the proposed analytics system can enhance the quality of educational services for students in an innovative e-learning framework.
文摘With this work, we introduce a novel method for the unsupervised learning of conceptual hierarchies, or concept maps as they are sometimes called, which is aimed specifically for use with literary texts, as such distinguishing itself from the majority of research literature on the topic which is primarily focused on building ontologies from a vast array of different types of data sources, both structured and unstructured, to support various forms of AI, in particular, the Semantic Web as envisioned by Tim Berners-Lee. We first elaborate on mutually informing disciplines of philosophy and computer science, or more specifically the relationship between metaphysics, epistemology, ontology, computing and AI, followed by a technically in-depth discussion of DEBRA, our dependency tree based concept hierarchy constructor, which as its name alludes to, constructs a conceptual map in the form of a directed graph which illustrates the concepts, their respective relations, and the implied ontological structure of the concepts as encoded in the text, decoded with standard Python NLP libraries such as spaCy and NLTK. With this work we hope to both augment the Knowledge Representation literature with opportunities for intellectual advancement in AI with more intuitive, less analytical, and well-known forms of knowledge representation from the cognitive science community, as well as open up new areas of research between Computer Science and the Humanities with respect to the application of the latest in NLP tools and techniques upon literature of cultural significance, shedding light on existing methods of computation with respect to documents in semantic space that effectively allows for, at the very least, the comparison and evolution of texts through time, using vector space math.
文摘The increased ownership of mobile phone users and the advancement of mobile applications enlarge the practicality and popularity of use for learning purposes among Chinese university students.However,even if innovative functions of these applications are increasingly reported in relevant research in the education field,little research has been in the application of spoken English language.This paper examined the effect of using a Mobile-Assisted Language Learning(MALL)application“IELTS Liulishuo”(speaking English fluently in the IELTS test)as a unit of analysis to improve the English-speaking production of university students in China.The measurement of this mobile application in its effectiveness of validity and reliability is through the use of seven dimensional criteria.Although some technical and pedagogical issues challenge adoptions of MALL in some less-developed regions in China,the study showed positive effects of using a MALL oral English assessment application characterised with Automatic Speech Recognition(ASR)system on the improvement of complexity,accuracy,and fluency of English learners in China’s colleges.
文摘目的开发一种深度学习系统用于成人发育性髋关节发育不良(Developmental dysplasia of the hip,DDH)患者的Crowe分型辅助诊断,并且分析该系统对于帮助临床医学生掌握DDH分型的可行性。方法纳入149例X线片训练集、42例测试集以及21例验证集,分割盆骨、提取DDH局部图像块,将金标准结果与医学生、AI辅助医学生评估结果进行比较。结果测试集共纳入42例,其中女性30例,男性12例,年龄(69±12)岁,涉及发育不良髋关节67侧(左30侧,右37侧)。AI、医学生、AI辅助医学生评估结果与金标准的相关性为0.906[95%CI(0.850,0.941)]、0.823[95%CI(0.726,0.887)]、0.886[95%CI(0.821,0.929)];准确率分别为0.87、0.78、0.88;精确度分别为0.88、0.83、0.89;召回率分别为0.87、0.78、0.88;F1值分别为0.87、0.80、0.88。混淆矩阵和条件概率结果显示,预测准确率Ⅰ型DDH三组分别为0.98、0.88、0.96,Ⅱ型DDH三组分别为0.40、0.20、0.40,Ⅲ型DDH三组分别为0.56、0.67、0.78;Ⅳ型DDH三组分别为0.88、0.75、0.88。结论深度学习辅助诊断系统可以有效提高医学生对于DDH分型的评估能力,可作为医学生学习掌握DDH影像诊断的培训工具。
