While modelling students’ learning behavior or preferences has been found as a crucial indicator for their course achievement, very few studies have considered it in predicting achievement of ...students in online courses. This study aims to model students’ online learning behavior and accordingly predict their course achievement. First, feature vectors are developed using their aggregated action logs during a course. Second, some of these feature vectors are quantified into three numeric values that are used to model students’ learning behavior, namely, accessing learning resources (content access), engaging with peers (engagement), and taking assessment tests (assessment). Both students’ feature vectors and behavior model constitute a comprehensive students’ learning behavioral pattern which is later used for prediction of their course achievement. Lastly, using a multiple criteria decision-making method (i.e., TOPSIS), the best classification methods were identified for courses with different sizes. Our findings revealed that the proposed generalizable approach could successfully predict students’ achievement in courses with different numbers of students and features, showing the stability of the approach. Decision Tree and AdaBoost classification methods appeared to outperform other existing methods on different datasets. Moreover, our results provide evidence that it is feasible to predict students’ course achievement with a high accuracy through modelling their learning behavior during online courses.
Recommendation systems are very important in various applications and e-commerce environments. A representative method is collaborative filtering (CF), which models the user preference by means of ...feedback from the user. CF-based methods made better recommendations than did previous studies because CF captures the interactions between the user and the item. However, despite the advantages of working with high-density data, these methods are vulnerable to the data sparsity that often exists in real data sets. In addressing this issue, we combine similarity-based approaches (which clearly serve product recommendations that are similar products) with knowledge-based similarity and provide individualized top-N recommendations. This approach, called UK (Unifying user preference and item knowledge-based similarity models), further exploits knowledge-based similarity ideas along with user preferences to extend the item interactions. We assume strong independence between various factors. We quantitatively demonstrate that by applying our method to real data sets of various sizes or types, UK works better than cutting-edge methods. In terms of qualitative discovery, UK also understands individual interactions and can provide meaningful recommendations according to the goal.
Pre-processing and post-processing are significant aspects of natural language processing (NLP) application software. Pre-processing in neural machine translation (NMT) includes subword tokenization ...to alleviate the problem of unknown words, parallel corpus filtering that only filters data suitable for training, and data augmentation to ensure that the corpus contains sufficient content. Post-processing includes automatic post editing and the application of various strategies during decoding in the translation process. Most recent NLP researches are based on the Pretrain-Finetuning Approach (PFA). However, when small and medium-sized organizations with insufficient hardware attempt to provide NLP services, throughput and memory problems often occur. These difficulties increase when utilizing PFA to process low-resource languages, as PFA requires large amounts of data, and the data for low-resource languages are often insufficient. Utilizing the current research premise that NMT model performance can be enhanced through various pre-processing and post-processing strategies without changing the model, we applied various decoding strategies to Korean–English NMT, which relies on a low-resource language pair. Through comparative experiments, we proved that translation performance could be enhanced without changes to the model. We experimentally examined how performance changed in response to beam size changes and n-gram blocking, and whether performance was enhanced when a length penalty was applied. The results showed that various decoding strategies enhance the performance and compare well with previous Korean–English NMT approaches. Therefore, the proposed methodology can improve the performance of NMT models, without the use of PFA; this presents a new perspective for improving machine translation performance.
Translation of the languages of ancient times can serve as a source for the content of various digital media and can be helpful in various fields such as natural phenomena, medicine, and science. ...Owing to these needs, there has been a global movement to translate ancient languages, but expert minds are required for this purpose. It is difficult to train language experts, and more importantly, manual translation is a slow process. Consequently, the recovery of ancient characters using machine translation has been recently investigated, but there is currently no literature on the machine translation of ancient Korean. This paper proposes the first ancient Korean neural machine translation model using a Transformer. This model can improve the efficiency of a translator by quickly providing a draft translation for a number of untranslated ancient documents. Furthermore, a new subword tokenization method called the Share Vocabulary and Entity Restriction Byte Pair Encoding is proposed based on the characteristics of ancient Korean sentences. This proposed method yields an increase in the performance of the original conventional subword tokenization methods such as byte pair encoding by 5.25 BLEU points. In addition, various decoding strategies such as n-gram blocking and ensemble models further improve the performance by 2.89 BLEU points. The model has been made publicly available as a software application.
The aim of a spelling correction task is to detect spelling errors and automatically correct them. In this paper we aim to perform the Korean spelling correction task from a machine translation ...perspective, allowing it to overcome the limitations of cost, time and data. Based on a sequence to sequence model, the model aligns its source sentence with an ‘error filled sentence’ and its target sentence aligned to the correct counter part. Thus, ‘translating’ the error sentence to a correct sentence. For this research, we have also proposed three new data generation methods allowing the creation of multiple spelling correction parallel corpora from just a single monolingual corpus. Additionally, we discovered that applying the Copy Mechanism not only resolves the problem of overcorrection but even improves it. For this paper, we evaluated our model upon these aspects: Performance comparisons to other models and evaluation on overcorrection. The results show the proposed model to even out-perform other systems currently in commercial use.
Artificial neural networks (ANNs) have proven to be among the most important artificial intelligence (AI) techniques in educational applications, providing adaptive educational services. However, ...their educational potential is limited in practice due to challenges such as the following: (i) the difficulties in incorporating symbolic educational knowledge (e.g., causal relationships and practitioners’ knowledge) in their development, (ii) a propensity to learn and reflect biases, and (iii) a lack of interpretability. As education is classified as a ‘high-risk’ domain under recent regulatory frameworks like the EU AI Act—highlighting its influence on individual futures and discrimination risks—integrating educational insights into ANNs is essential. This ensures that AI applications adhere to essential educational restrictions and provide interpretable predictions. This research introduces NSAI, a neural-symbolic AI approach that integrates neural networks with knowledge representation and symbolic reasoning. It injects and extracts educational knowledge into and from deep neural networks to model learners’ computational thinking, aiming to enhance personalized learning and develop computational thinking skills. Our findings revealed that the NSAI approach demonstrates better generalizability compared to deep neural networks trained on both original training data and data enriched by SMOTE and autoencoder methods. More importantly, we found that, unlike traditional deep neural networks, which mainly relied on spurious correlations in their predictions, the NSAI approach prioritizes the development of robust representations that accurately capture causal relationships between inputs and outputs. This focus significantly reduces the reinforcement of biases and prevents misleading correlations in the models. Furthermore, our research showed that the NSAI approach enables the extraction of rules from the trained network, facilitating interpretation and reasoning during the path to predictions, as well as refining the initial educational knowledge. These findings imply that neural-symbolic AI not only overcomes the limitations of ANNs in education but also holds broader potential for transforming educational practices and outcomes through trustworthy and interpretable applications.
Data mining approaches have proven to be successful in improving learners' interaction with educational computer games. Despite the potential of predictive modelling in providing timely adaptive ...learning and gameplay experience, there is a lack of research on the early prediction of learners' performance in educational games. In this research, we propose an early predictive modelling approach, called GameEPM, to estimate learners' final scores in an educational game for promoting computational thinking. Specifically, the GameEPM approach models the sequence of learners' actions and then uses a limited sequence of the actions to predict the final score of the game for each learner. The findings from our initial trials show that our approach can accurately and robustly estimate the learners' performance at the early stages of the game. Using less than 50% of learners' action sequences, the cross-validated deep learning model achieves a squared correlation higher than 0.8 with a relative error of less than 8%, outperforming a range of regression models like linear regression, random forest, neural networks, and support vector machines. An additional experiment showed that the validated deep learning model can also achieve high performance while tested on an independent game dataset, showing its applicability and robustness in real-world cases. Comparing the results with traditional machine learning methods revealed that, in the validation and application phases, up to 0.30 and 0.35 R2 gain is achieved in favor of the deep learning model, respectively. Finally, we found that while the lengths of action sequences influence the predictive power of the traditional machine learning methods, this effect is not substantial in the deep learning model.
Grammar error correction (GEC) refers to the proper correction of grammatical errors in a given sentence. Important factors to consider in GEC are not only the grammatical correction of the sentence, ...but also the recognition of a correct sentence in which no changes are required. However, GEC approaches in which deep learning recently started being used consider only the former aspect, which leads to overcorrection, whereby changes are made to a correct sentence unnecessarily. Because this bias is also reflected in performance metrics, conventional performance metrics consider only part of the important factors in GEC. This study proposes a new metric to consider both important aspects in GEC and to provide a new viewpoint for the GEC task. To the best of the authors knowledge, this study is the first to deal with comprehensively considering the correction performance and overcorrection problem in GEC. The experimental results demonstrate that the model performance ranking was reversed when evaluating the performance with the proposed metric compared to the General Language Understanding Evaluation benchmark <xref ref-type="bibr" rid="ref21">21 , which only considers the correction performance. This indicates that the high performance of the correction does not result in less problems with the overcorrection and that the overcorrection problem should also be considered when evaluating the model performance. Moreover, we found that the copy mechanism <xref ref-type="bibr" rid="ref14">14 helps to alleviate the problem of overcorrection.
This research aims to explore the comprehension of historical Korean archives authored by common literati. Numerous endeavors have been made to study Korean historical documents; however, the ...majority of these endeavors focus solely on royal documents. By comparing the distinct linguistic characteristics between royal and commoner languages, this study challenges the applicability of the royal language-centric approach to commoner documents. In particular, we investigate the feasibility and limitations of existing resources that share the same writing system (Hanja) as historical Korean documents for processing Korean common literati documents. Through our investigation, we propose a simple yet effective methodology that enables the utilization of Hanja-based language resources in processing Korean common literati documents: the removal of special characters. We demonstrate that aligning characteristics of Hanja-based resources allows considerable performance improvements. To the best of our knowledge, our study represents the first research endeavor to concentrate on the comprehension of common literati documents.
Predictive learner modelling is crucial for personalized education. While convolutional neural networks (CNNs) have shown great success in education, their potential in learner modelling via image ...data is unexplored. This research introduces a novel and interpretable approach for Image-based Learner Modelling (ImageLM) using CNNs and transfer learning to model learners’ performance and accordingly classify their computational thinking solutions. The approach integrates Grad-CAM, enabling it to provide insights into its decision-making process. Findings show that our custom CNN outperforms other models (namely ResNet, VGG, and Inception), with 83% accuracy in predicting solution correctness. More importantly, the ImageLM approach identifies the regions that contribute the most to the predictions, shedding light on learners' computational thinking knowledge and advancing toward trustworthy AI for education. These results underline the potential of utilizing imagery data from learners’ activities during the learning process to predict their performance, especially in challenging environments like programming where traditional feature extraction and learning might struggle.