Search

Abstract In recent years, smart learning environments have become central to modern education and support students and instructors through tools based on prediction and recommendation models. These methods often use learning material metadata, such as the knowledge contained in an exercise which is usually labeled by domain experts and is costly and difficult to scale. It recognizes that automated labeling eases the workload on experts, as seen in previous studies using automatic classification algorithms for research papers and Japanese mathematical exercises. However, these studies didn’t delve into fine-grained labeling. In addition to that, as the use of materials in the system becomes more widespread, paper materials are transformed into PDF formats, which can lead to incomplete extraction. However, there is less emphasis on labeling incomplete mathematical sentences to tackle this problem in the previous research. This study aims to achieve precise automated classification even from incomplete text inputs. To tackle these challenges, we propose a mathematical exercise labeling algorithm that can handle detailed labels, even for incomplete sentences, using word n-grams, compared to the state-of-the-art word embedding method. The results of the experiment show that mono-gram features with Random Forest models achieved the best performance with a macro F-measure of 92.50%, 61.28% for 24-class labeling and 297-class labeling tasks, respectively. The contribution of this research is showing that the proposed method based on traditional simple n-grams has the ability to find context-independent similarities in incomplete sentences and outperforms state-of-the-art word embedding methods in specific tasks like classifying short and incomplete texts.

In the realm of mathematics education, self-explanation stands as a crucial learning mechanism, allowing learners to articulate their comprehension of intricate mathematical concepts and strategies. As digital learning platforms grow in prominence, there are mounting opportunities to collect and utilize mathematical self-explanations. However, these opportunities are met with challenges in automated evaluation. Automatic scoring of mathematical self-explanations is crucial for preprocessing tasks, including the categorization of learner responses, identification of common misconceptions, and the creation of tailored feedback and model solutions. Nevertheless, this task is hindered by the dearth of ample sample sets. Our research introduces a semi-supervised technique using the large language model (LLM), specifically its Japanese variant, to enrich datasets for the automated scoring of mathematical self-explanations. We rigorously evaluated the quality of self-explanations across five datasets, ranging from human-evaluated originals to ones devoid of original content. Our results show that combining LLM-based explanations with mathematical material significantly improves the model’s accuracy. Interestingly, there is an optimal limit to how many synthetic self-explanation data can benefit the system. Exceeding this limit does not further improve outcomes. This study thus highlights the need for careful consideration when integrating synthetic data into solutions, especially within the mathematics discipline.

This paper explores co-design in Japanese education for deploying data-driven educational technology and practice. Although there is a growing emphasis on data to inform educational decision-making and personalize learning experiences, challenges such as data interoperability and inconsistency with teaching goals prevent practitioners from participating. Co-design, characterized by involving various stakeholders, is instrumental in addressing the evolving needs of technology deployment. Japan's educational context aligns with co-design implementation, with a learning and evidence analytics infrastructure facilitating data collection and analysis. From the Japanese co-design practice of educational technologies, the paper highlights a 6-phase co-design framework: motivate, pilot, implement, refine, evaluate, and maintain. The practices focus on data-driven learning strategies, technology interventions, and across-context dashboards, covering assorted learning contexts in Japan. By advocating for a co-design culture and data-driven approaches to enhance education in Japan, we offer insights for education practitioners, policymakers, researchers, and industry developers.

Many modern learning systems rely on a data representation of the knowledge that is to be learned to estimate a learner's mastery state and recommend appropriate learning tasks to further improve their acquisition of knowledge and skills. In particular, the rapid development of intelligent tutoring systems (ITS) and standardized curricula has increased the need for information on knowledge structures and their links to learning materials and tasks. However, manually labeling educational data has traditionally been a time-consuming, labor-intensive task, and thus has limited its use by time-constrained teachers and practitioners. In previous research, a range of machine-learning methods have been proposed to address this problem, with only a few of them focusing on Japanese educational datasets from secondary schools. In this paper, to support the labeling of Japanese mathematics exercises by teachers and other domain experts, we apply natural language processing techniques including word-embedding and key-phrase-based exercise-to-exercise similarity methods. We evaluated the proposed method by both the performance of the models when compared to several state-of-the-art methods, and also its effectiveness in supporting humans in the task of labeling educational materials. Through this two-phase evaluation, we found that the proposed method outperformed other methods, and when implemented in a human-in-the-loop system it achieved significantly more accuracy and consumed less time for the task of labeling mathematics exercises. [ABSTRACT FROM AUTHOR]