Learner Model Research Articles

J48-Based Decision Tree Algorithm in Detecting Kolb’s Learning Style Preferences of Information Technology Students

Teacher education institutions developed interventions to augment both the Teacher’s teaching and the Learner’s learning process. Previous studies focused on detecting learning styles in e-learning using learning management systems and adaptive learning in the online learning process, specifically using the Felder-Silverman learning style model. On the contrary, this study aimed to develop a decision tree-based model to detect the Learner’s learning style inspired by Kolb’s learning style in a face-to-face learning environment. Knowledge discovery in databases through data mining was utilized using the J48 algorithm to develop a decision tree-based model. This study was participated by 408 out of 462 information technology students in a state university in the Philippines. The study’s result was able to develop four J48-based decision trees with conditional rule models for activist, reflector, theorist and pragmatist learners. The evaluation of the decision tree models using confusion matrix and receiver operating curve showed a very high accuracy detection of every learning style (weighted average of 88-96%). This result recommends applying this in an actual system or computer application for easy and fast learning style detection based on the characteristics of the learners.

Explainable machine learning-aided efficient prediction model and software tool for bond strength of concrete with corroded reinforcement

The bond strength between concrete and reinforcement is crucial for the composite action and serviceability of reinforced concrete (RC) structures. However, it is vulnerable to deterioration from the corrosion of reinforcement bars, especially in marine structures. Thus, a precise and reliable model for the bond strength in corrosive environments is necessary to evaluate the serviceability and structural performance of corroded RC members. This study employs explainable machine learning (ML) techniques to assess the bond strength between concrete and corroded bars. Eight ML models are developed to establish the best predictive model for bond behavior, considering seven input parameters: corrosion level (CL), steel yield strength, compressive strength of concrete, concrete cover-to-bar diameter ratio, bar diameter-to-bonded length ratio, reinforcement type, and test type. The super learner (SL) model, integrating three ML models, outperforms other models and analytical methods with a large R2 value (98% on the test set) and minimal statistical errors. The SHapley Additive exPlanation (SHAP) technique identifies CL as the most influential parameter on bond strength, while the reinforcement and test types have the least effect. Finally, a user-friendly graphical user interface (GUI) tool is established to facilitate the practical implementation of the developed model and support accurate bond strength prediction in concrete with steel reinforcement under corrosive environments.

Design and optimization of an open personalized human-computer interaction system for New Year Painting based on the learner's model

Design and optimization of an open personalized human-computer interaction system for New Year Painting based on the learner's model

A model for teaching students to create educational games in supplementary education in computer science

Relevance. The relevance of the stated topic of this research is determined by the presence of significant changes in modern educational processes, which have caused an increase in the importance of issues of digital literacy and information technology education for students at all levels of educational institutions. Purpose. The main aim of this research is to explore the key principles of a model for teaching students how to create educational games in further education in computer science, with the programming environment Scratch selected as an example. Methodology. The methodological approach in this research is based on a combination of the method of system analysis of the principles of using educational computer games in the educational process and a comprehensive study of the impact of game-based methods of teaching classes on students’ learning of educational material. The scientific research was conducted on the basis of the Kazakh National Pedagogical University named after Abay. Results. In this scientific study the results were obtained, indicating the existence of fundamental opportunities for students to prepare individual models of learning games aimed at stimulating their activity in the learning process, increasing their motivation and involvement in the learning process. In doing so, educational games can be used to create a quality model for the individual learner; in fact, such games provide ample opportunity for the learner to interact with the computer, which can be used in the future to develop effective educational game models when developing additional teaching methods in computer science. Conclusions. The practical relevance of the results obtained in this research study and the conclusions formulated on their basis lies in the possibility of their application in the creation of additional game learning materials designed to develop the computer science competencies of students acquired during their studies in an educational institution.

THE INFLUENCE OF PICTURE AND PICTURE TYPE COOPERATIVE LEARNING MODELS ON IPS LEARNING OUTCOMES

We can know the learning model from various sources as material for teaching and learning activities, in fact, the learning model exists to help teachers in the teaching and learning process so that it can be conveyed to students better and not saturated when the teaching and learning process takes place. Picture and picture type cooperative learning model as one of the learning models that can be used in the process of teaching and learning activities, where this model uses pictures as teaching media, but in fact, social studies educators, especially at SMP Negeri 2 Leles, rarely use other learner models and only use conventional models as learning models in the process of teaching and learning activities so that students tend to get bored, making many students' learning outcomes still under KKM. To determine the effect of learning using this picture and picture type cooperative learning model, researchers use a quantitative approach with a type of quasi-experimental research. The subjects of the study were class VIII SMP Negeri 2 Lelesa with class VIIIA as the experimental class and class VIIIB as the control class. The results showed differences in student learning outcomes before and after using the picture and picture type cooperative learning model in the teaching and learning process, this can be seen from the post-test scores that exceed the pre-test scores, this shows that students better understand social studies learning material when using learning models in the teaching and learning process, meaning that the picture and picture type cooperative learning model has a good influence on social studies learning outcomes where students experience improved learning outcomes.

Enhancing Academic Outcomes through an Adaptive Learning Framework Utilizing a Novel Machine Learning-Based Performance Prediction Method

Introduction:E landscapes have been transformed by technological advancements, enabling adaptive and flexible learning through AI-based and decision-oriented adaptive learning systems. The increasing importance of this solutions is underscored by the pivotal role of the learner model, representing the core of the teaching-learning dynamic. This model, encompassing qualities, knowledge, abilities, behaviors, preferences, and unique distinctions, plays a crucial role in customizing the learning experience. It influences decisions related to learning materials, teaching strategies, and presentation styles. Objective: This study meets the need for applying AI-driven adaptive learning in education, implementing a novel method that uses self-esteem (ES), emotional intelligence (EQ), and demographic data to predict student performance and adjust the learning process. Methods: Our study involved collecting and processing data, constructing a predictive machine learning model, implementing it as an online solution, and conducting an experimental study with 146 high school students in computer science and French as foreign language. The aim was to tailor the teaching-learning process to the learners' performance. Results: significant correlations were observed between self-esteem, emotional intelligence, demographic data, and final grades. The predictive model demonstrated a 90 % accuracy rate. In the experimental group, the results indicated higher scores, with an average of 15,78/20 compared to the control group's 12,53/20 in computer science. Similarly, in French as a foreign language, the experimental group achieved an average of 13,78/20, surpassing the control group's 10,47/20. Conclusion: the achieved results motivate the creation of a multifactorial AI-driven adaptive learning platform. Recognizing the necessity for improvement, we aim to refine the predicted performance score through the incorporation of a diagnostic evaluation, ensuring an optimal grouping of learners

Solving the imbalanced data issue: automatic urgency detection for instructor assistance in MOOC discussion forums

In MOOCs, identifying urgent comments on discussion forums is an ongoing challenge. Whilst urgent comments require immediate reactions from instructors, to improve interaction with their learners, and potentially reducing drop-out rates—the task is difficult, as truly urgent comments are rare. From a data analytics perspective, this represents a highly unbalanced (sparse) dataset. Here, we aim to automate the urgent comments identification process, based on fine-grained learner modelling—to be used for automatic recommendations to instructors. To showcase and compare these models, we apply them to the first gold standard dataset for Urgent iNstructor InTErvention (UNITE), which we created by labelling FutureLearn MOOC data. We implement both benchmark shallow classifiers and deep learning. Importantly, we not only compare, for the first time for the unbalanced problem, several data balancing techniques, comprising text augmentation, text augmentation with undersampling, and undersampling, but also propose several new pipelines for combining different augmenters for text augmentation. Results show that models with undersampling can predict most urgent cases; and 3X augmentation + undersampling usually attains the best performance. We additionally validate the best models via a generic benchmark dataset (Stanford). As a case study, we showcase how the naïve Bayes with count vector can adaptively support instructors in answering learner questions/comments, potentially saving time or increasing efficiency in supporting learners. Finally, we show that the errors from the classifier mirrors the disagreements between annotators. Thus, our proposed algorithms perform at least as well as a ‘super-diligent’ human instructor (with the time to consider all comments).

Temporal learner modelling through integration of neural and symbolic architectures

Temporal learner modelling through integration of neural and symbolic architectures

An ensemble of CNNs with self-attention mechanism for DeepFake video detection

The availability of large-scale facial datasets with the rapid progress of deep learning techniques, such as Generative Adversarial Networks, has enabled anyone to create realistic fake videos. These fake videos can potentially become harmful when used for fake news, hoaxes, and identity fraud. We propose a deep learning bagging ensemble classifier to detect manipulated faces in videos. The proposed bagging classifier uses the convolution and self-attention network (CoAtNet) model as a base learner. CoAtNet model is vertically stacking depthwise convolution layers and self-attention layers in such a way that generalization, capacity, and efficiency are improved. Depthwise convolution captures local features from faces extracted from video then pass these features to the attention layers to extract global information and efficiently capture long-range dependencies of spatial details. Each learner is trained on a different subset randomly taken of training data with a replacement then models’ predictions are combined to classify the video either as real or fake. We also use CutMix data augmentation on the extracted faces to enhance the generalization and localization performance of the base learner model. Our experimental results show that our proposed method achieves higher efficiency compared to state-of-the-art methods with AUC values of 99.70%, 97.49%, 98.90%, and 87.62% on the different manipulation techniques of the FaceForensics++ dataset (DeepFakes (DF), Face2Face (F2F), FaceSwap (FS), and NeuralTextures (NT)), respectively, and 99.74% on the Celeb-DF dataset.

AMAD: Active learning-based multivariate time series anomaly detection for large-scale IT systems

Multivariate time series anomaly detection on key performance indicators helps mitigate the impact of large-scale IT system anomalies. Due to the large volume and the abstract nature of multivariate time series, previous works have tended to make overly strict or optimistic hypotheses on labeling costs and resulted in unsatisfactory results. Thus, it remains a challenge to make an appropriate trade-off between labeling costs and model performance. This research proposes AMAD, an active learning-based approach that works to address this problem. Its core idea is to provide the learner model high-value label queries via an ensemble query strategy, which dynamically adapts to the estimated anomaly ratio and the ever-changing model performance. Moreover, it is the first to discuss in detail the margin effect of query strategies on model performance, to our knowledge, this has not been investigated in previous works on time series anomaly detection. Extensive experiments on five public datasets demonstrate that AMAD works well and robustly on various real-world scenarios, which outperforms state-of-the-art baseline methods by 16% and 11% in terms of recall and F1, with only 3% of data being labeled.

Improving road safety with ensemble learning: Detecting driver anomalies using vehicle inbuilt cameras

The adoption of Advanced Driver Assistance Systems (ADAS) has expanded dramatically in recent years, with the goal of improving road safety and driving comfort. Driver monitoring is important to ADAS since it identifies abnormalities such as sleepiness, distraction, and impairment to guarantee safe vehicle operation. Traditional methods of detecting driver anomalies rely on intrusive physiological measures, while ADAS with built-in cameras offers a non-intrusive and cost-effective option. This study investigates the application of ensemble model learning for driver anomaly detection in automobiles employing ADAS and in-vehicle cameras. Deep learning models such as ResNet50, DenseNet201, and Inception V3 were deployed as learner models to classify driving behavior. The raw dataset used in this study was in the form of videos obtained from the National Tsinghua Driver Drowsiness Detection (NTHUDD) dataset. Amongst the two ensemble models used, the eXtreme Gradient Boost (XGBoost) classifier pooled predictions from the learner models. It attained a remarkable average accuracy and precision of 99% on the validation dataset. Classes such as laugh_talk and yawning were properly and separately distinguished. The ensemble technique capitalized on the strengths of various models while mitigating their weaknesses, resulting in robust and trustworthy forecasts. The findings highlight the potential of ensemble modeling to enhance driver anomaly detection systems, providing valuable insights for improving road safety. By continually monitoring driver behavior and detecting abnormalities, ADAS can provide timely warnings and interventions to prevent accidents and save human lives.

A Review of Learner’s Model for Programming in Teaching and Learning

Over recent years, computer science (CS) teachers and instructor have faced several challenges in assisting students to strengthen their understanding of programming. To make matters worse, existing assessment methods could be more effective in assessing students' programming skills and knowledge, thus entailing a review of issues surrounding the instructions of programming courses. Against this backdrop, the authors accomplished a systematic review of the current literature to identify several socio-cognitive factors that can help develop a learner model for learning programming. Specifically, the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) technique was utilized to identify and select pertinent articles from three primary online databases: Scopus, Web of Science, and Eric. Initially, 401 relevant papers were identified and retrieved, which were further reduced to only 24 articles based on specific selection criteria. As revealed, several demographic factors (such as gender, age, ethnicity, and SES) and socio-cognitive factors (motivation, attitude, and interest) have been shown to wield significant impacts on student learning of programming. The authors' findings of the systematic literature review helped synthesize the learner model's essential elements that must be carefully considered and used. Arguably, the use of such a new learner model can oblige guide instructors to teach programming more efficaciously by crystallizing several students' socio-cognitive backgrounds, which collectively have a significant impact on student learning of programming courses or subjects at the primary, secondary, and tertiary levels of education, especially in the Malaysian educational context.

Application of University English Teaching Model Based on Artificial Intelligence Technology

Abstract This paper is based on cloud computing technology in artificial intelligence and a clustering algorithm controlled to divide teaching resource attributes and reduce packet loss rates in the system. Data mining is used to reduce the number of iterations and improve processing efficiency, a learner model is designed, and the teaching process and assessment tasks are constructed to realize the teaching model. Two natural classes of students at the University of S. underwent an instructional experiment. The results showed that the experimental class reached about 70% of the overall students in the achievement range of 85-95 and 95-100. The college English teaching model based on artificial intelligence technology’s feasibility and effectiveness are proven by the significant improvement in performance compared to the control class.

ImageLM: Interpretable image-based learner modelling for classifying learners’ computational thinking

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.

The Application of Artificial Intelligence Teaching Software in College English Teaching

Abstract This paper first proposes English learning software based on personalized recommendations to enhance students’ interest and motivation in learning college English based on artificial intelligence teaching. Secondly, it consists of constructing a learner model, an English learning resource personalized recommendation algorithm to constitute an English learning resource recommendation model, and using a collaborative decomposition algorithm and a regularization algorithm to integrate the potential, multifaceted trust relationship between users with the traditional scoring matrix, a process that emphasizes that trust between users in English teaching varies by different factors and strengths. Then, a personalized recommendation-based university English learning software was designed, and English learning software applications were analyzed by example. The results showed that the adjusted means of pre and post-test scores of the control group, experimental group one, and experimental group two were 77.37, 80.94, and 75.12, respectively, and the statistical score result F=5.33, P<05 indicated that there was a significant difference in the change of English performance of these three groups, which verified that the teaching software had a facilitating effect on the improvement of English teaching quality.

Analysis of metabolites in human gut: illuminating the design of gut-targeted drugs

Gut-targeted drugs provide a new drug modality besides that of oral, systemic molecules, that could tap into the growing knowledge of gut metabolites of bacterial or host origin and their involvement in biological processes and health through their interaction with gut targets (bacterial or host, too). Understanding the properties of gut metabolites can provide guidance for the design of gut-targeted drugs. In the present work we analyze a large set of gut metabolites, both shared with serum or present only in gut, and compare them with oral systemic drugs. We find patterns specific for these two subsets of metabolites that could be used to design drugs targeting the gut. In addition, we develop and openly share a Super Learner model to predict gut permanence, in order to aid in the design of molecules with appropriate profiles to remain in the gut, resulting in molecules with putatively reduced secondary effects and better pharmacokinetics.

Grade 6 primary school learners’ views on minibus taxis and their drivers

Background: This study identifies potential assets of the minibus taxi industry space and alerts taxi drivers to the expectations of various role-players in education, through understanding the views of various role-players about minibus taxi drivers and their role in supporting education.Aim: Viewing the minibus taxi industry through alternative lenses may lead to development which may ultimately lead to an improvement in the attitudes of the taxi industry towards the education sector and perceived negative behaviour of taxi drivers.Setting: The study took place at three schools in Gauteng, South Africa.Methods: The study followed an interpretive, qualitative research approach for exploring and understanding various views. A case study design using Lefebvre’s, ‘Production of Space’ theory and asset-based theory, were used as a theoretical framework in conjunction with the Johari Window model. Convenient sampling was applied. Focus group interviews, semi-structured questionnaires, fieldnotes, and observation were used for data collection and construction.Results: The results indicated that views which are embedded in individuals have many facets; hence, it is important to be aware of them to be able to identify the assets of minibus taxi drivers.Conclusion: Minibus taxi drivers are key role players, valuable facilitators and role models for learners, although educational development is needed.Contribution: Very little work has been done on the educational value of spaces such as taxis and transport vehicles; this research adds considerable value to this research lacuna.

Acquisition of English nominal suffix -er by advanced EFL learners: a view from usage-based perspective

Abstract The present study investigated advanced Croatian EFL learners’ knowledge of five meanings of the English nominal (deverbal) suffix -er. It probed their ability to comprehend and produce corpus-rare and presumably unentrenched -er nouns in their prototypical agent and instrument meanings and their non-prototypical patient, locative, and causative meanings. It was hypothesized that participants would deal effortlessly with agent and instrument meanings of the low-frequency nouns since the corpus-attested high type frequency of -er agents and instruments, among others, suggests the existence of productive corresponding schemas. We hypothesized that participants would struggle with patient, locative and causative meanings of the low-frequency nouns since the corpus-attested low type frequency of the three functions arguably does not support their association with -er. A recognition and a production test were administered to two separate groups of English majors at a Croatian public university (n = 131). Results confirm general usage-based predictions about better performance with low-frequency agent and instrument -er nouns. However, a detailed examination reveals unexpected results, which confirm that frequency, however important, is not the only factor to include in a future model of EFL learners’ derivational proficiency.

Could the Technology for Adaptive Learning Systems Come out of GBL?

Games based learning (GBL) has a related field that it can draw upon: intelligent tutoring systems (ITS). Acommon concern in both fields is adaptivity, whereby the system can automatically adapt to the user. In order to supportthis adaptation, an ITS will generally include a user model and may also have a formal domain model. Work in this areastarted optimistically some years ago, but seems to have either lost some of that initial enthusiasm or been diverted intoother directions. We scan recent ITS literature to help consider why this might be, and suggest how GBL may be the fieldbest placed to take the work forward again. Learner models have long been seen as useful for adaptive learning systems.They include information about the learner which allows the system to adapt the course of learning materials andexercises to the learner's particular characteristics. In order to achieve a good quality of adaptation to the user, a detailedmodel of the required domain knowledge is typically added. The user and domain models then have to be broughttogether to lay out a course of exercises for the learner to do, and to track progress as the knowledge is learned. It's anattractive research programme, but recent work has moved to new issues, such as MOOCs. The reasons for that are partlyopportunistic and economical, but also suggest a deeper problem with the research programme. It is a costly task todevelop a domain model, and a suitable learner model that can take advantage of it. We suggest that GBL is in a goodposition to push through this cost barrier, because much of the effort is already implicitly involved in the game designprocess, which typically has to be more rigorously planned out in order to make the game a good one. One might thusexpect the next breakthroughs in adaptive learning systems to come from GBL. We further argue that the advantages toresearch, offered by the ITS framework, are also potentially beneficial to the way we teach the subject of GBL to ourstudents on game development courses.

Modeling of Distance Learners’ Retention Using Mixed-Model on Non-Proportional Hazard

Universitas Terbuka is a public university in Indonesia that implements an open-distance education system. The quality and success of higher education in Indonesia are indicated by the gross enrolment rate (GER). Modelling student retention is one indicator of the value of the GER. The chances of distance learners’ retention in a certain period can be determined through survival analysis, namely the “Cox model” or “Cox proportional hazard model”. The limitations of the Cox model pose new problems in modelling that involve more than two types of covariates and the presence of random effects. The problem that arises is the non-proportional hazard. The cause of non-proportional hazard is time-dependent covariates, so that individual risk changes over time. Another reason is the presence of random effects, both observed and unobserved. In this model, there are two random effects, namely, the unobserved random effect and the observed random effect. Thus, the model is called the mixed effect model of non-proportional hazard. Based on the characteristics of the covariates observed, appropriate survival analysis for modelling distance learners’ retention is a mixed model on the non-proportional hazard. The covariates involved in very complex modelling include time-independent covariates, time-dependent covariates, and unobserved random effects. The results of the analysis with the mixed model on non-proportional hazard showed that the covariates that significantly affected the distance learner’s retention were: educational background, age, grade point average, marital status, scholarship, independent learning culture, number of credits taken, and number of courses taken each semester.