Search

Abstract In inclusive education, students with different needs learn in the same context. With the advancement of artificial intelligence (AI) technologies, it is expected that they will contribute further to an inclusive learning environment that meets the individual needs of diverse learners. However, in Japan, we did not find any studies exploring current needs in an actual special needs context. In this study, we used the learning and evidence analysis framework (LEAF) as a learning analytics-enhanced learning environment and employed Active Reading as an example learning task to investigate the challenges and possibilities of applying AI to inclusive education in the future. Two students who attended a resource room formed the context. We investigated learning logs in the LEAF system while each student executed a given learning task. We detected specific learning behaviors from the logs and explored the challenges and future potential of learning with AI technology, considering human involvement in orchestrating inclusive educational practices.

Abstract The development of technology enables diverse learning experiences nowadays, which shows the importance of learners’ self-regulated skills at the same time. Particularly, the ability to allocate time properly becomes an issue for learners since time is a resource owned by all of them. However, they tend to struggle to manage their time well due to the lack of awareness of its existence. This study, hence, aims to reveal how learners allocate their time and evaluate the effectiveness of the time allocation by examining its effects on learners’ performance. We collect the learning logs of 116 seventh-graders from the online learning system implemented in a Japanese public junior high school. We look at the data in the time window of 34 days before the regular exam. Even though clustering techniques as a Learning Analytics method help identify different groups of learners, it is seldom applied to group students’ learning patterns with different levels of indicators extracted from their learning process data. In this study, we adopt the method to cluster students’ patterns of time allocation and find that better performance can result from the consistency of study time throughout the exam preparation period. Practical suggestions are then proposed for different roles involved in digital learning environments to facilitate students’ time management. Collectively, this study is expected to make contributions to smart learning environments supporting self-regulated learning in the digital era.

Abstract Digitized learning materials are a core part of modern education, and analysis of the use can offer insight into the learning behavior of high and low performing students. The topic of predicting student characteristics has gained a lot of attention in recent years, with applications ranging from affect to performance and at-risk student prediction. In this paper, we examine students reading behavior using a digital textbook system while taking an open-book test from the perspective of engagement and performance to identify the strategies that are used. We create models to predict the performance and engagement of learners before the start of the assessment and extract reading behavior characteristics employed before and after the start of the assessment in a higher education setting. It was found that strategies, such as: revising and previewing are indicators of how a learner will perform in an open ebook assessment. Low performing students take advantage of the open ebook policy of the assessment and employ a strategy of searching for information during the assessment. Also compared to performance, the prediction of overall engagement has a higher accuracy, and therefore could be more appropriate for identifying intervention candidates as an early-warning intervention system.

Abstract Blended learning technique has adapted many new digitized tools to facilitate students with flexible and self-phased learning opportunities. The flipped classroom strategy, one of the blended learning models has been limited by low engagement of students in the online component. In the present study, we augment a Flip and Pair (F&P), an active-learning strategy into the blended learning course. Following the AB type single group quasi-experimental design, we evaluated the effects of F&P strategy on the student’s engagement and learning while orchestrating it for an undergraduate engineering physics course. Our results highlighted that there is a positive correlation between the engagement (computed based on learning logs of TEEL (Technology-enhanced and Evidence-based Education and Learning) platform in the F&P activities with that of the performance score (knowledge quizzes and final exam). F&P strategy had a better contribution compared to Flip and Traditional Teaching (F&TT) strategy with respect to both engagement and performance. Also, students exhibited a positive perception of learning and engagement. Based on our findings, we identified the key instructional measures that an instructor can follow to increase student engagement while using the F&P strategy.

Abstract Early prediction systems have already been applied successfully in various educational contexts. In this study, we investigated developing an early prediction system in the context of eBook-based teaching-learning and used students’ eBook reading data to develop an early warning system for students at-risk of academic failure -students whose academic performance is low. To determine the best performing model and optimum time for possible interventions we created prediction models by using 13 prediction algorithms with the data from different weeks of the course. We also tested effects of data transformation on prediction models. 10-fold cross-validation was used for all prediction models. Accuracy and Kappa metrics were used to compare the performance of the models. Our results revealed that in a sixteen-week long course all models reached their highest performance with the data from the 15th week. On the other hand, starting from the 3rd week, the models classified low and high performing students with an accuracy of over 79%. In terms of algorithms, Random Forest (RF) outperformed other algorithms when raw data were used, however, with the transformed data J48 algorithm performed better. When categorical data were used, Naive Bayes (NB) outperformed other algorithms. Results also indicated that models with transformed data performed lower than the models created using categorical data. However, models with categorical data showed similar performance with models with raw data. The implications of the results presented in this research were also discussed with respect to the field of Learning Analytics.