[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-seo-600513-105":3,"detail-sidebar-cat-0-en-105":85,"doc-detail-600513-en":134},{"code":4,"msg":5,"data":6},0,"ok",{"site_id":7,"language":8,"slug":9,"title":10,"keywords":11,"description":12,"schema_data":13,"social_meta":78,"head_meta":80,"extra_data":82,"updated_unix":84},105,"en","enhanced-predictive-performance-a-comparative-analysis-of-ml-and-dl-models-using-on-augmented-lms-interaction-data","Enhanced predictive performance: A comparative analysis of ML and DL models using on augmented LMS interaction data","","This study examines how learning management system (LMS) interaction logs can be leveraged to predict higher-education students’ academic progress and to flag those at risk of failing. It develops and compares multiple machine learning (ML) and deep learning (DL) models using an integrated design science research (DSR) approach, including data augmentation and balancing to address class imbalance. Results benchmark ten model types—such as random forest, decision tree, CNN, MLP, and LSTM—showing strong DL performance, with the LSTM-based model achieving the highest and nearly perfect accuracy, enabling timely, data-driven interventions.",{"@graph":14,"@context":77},[15,34,56],{"@type":16,"itemListElement":17},"BreadcrumbList",[18,23,27,31],{"item":19,"name":20,"@type":21,"position":22},"https://docshare.wps.com","Home","ListItem",1,{"item":24,"name":25,"@type":21,"position":26},"https://docshare.wps.com/document/","Document",2,{"item":28,"name":29,"@type":21,"position":30},"https://docshare.wps.com/document/research-report/","Research & Report",3,{"item":32,"name":10,"@type":21,"position":33},"https://docshare.wps.com/document/enhanced-predictive-performance-a-comparative-analysis-of-ml-and-dl-models-using-on-augmented-lms-interaction-data/600513/",4,{"url":32,"name":10,"@type":35,"image":36,"author":41,"headline":10,"publisher":44,"fileFormat":47,"inLanguage":8,"description":12,"dateModified":48,"datePublished":49,"encodingFormat":47,"isAccessibleForFree":50,"interactionStatistic":51},"DigitalDocument",{"url":37,"@type":38,"width":39,"height":40},"https://docshare.wps.com/thumbnails/enhanced-predictive-performance-a-comparative-analysis-of-ml-and-dl-models-using-on-augmented-lms-interaction-data/600513.png","ImageObject",300,407,{"name":42,"@type":43},"OmBimo","Person",{"url":19,"name":45,"@type":46},"DocShare","Organization","application/pdf","2026-10-30","2026-10-15",true,{"@type":52,"interactionType":53,"userInteractionCount":55},"InteractionCounter",{"@type":54},"ViewAction",6,{"@type":57,"mainEntity":58},"FAQPage",[59,65,69,73],{"name":60,"@type":61,"acceptedAnswer":62},"What is the primary goal of the study?","Question",{"text":63,"@type":64},"To develop and compare ML and DL models that use augmented LMS interaction data to identify higher-education students at risk of failing with high classification accuracy.","Answer",{"name":66,"@type":61,"acceptedAnswer":67},"How does the study handle class imbalance in the predictive models?",{"text":68,"@type":64},"It incorporates data augmentation and balancing techniques as part of an integrated DSR methodology to improve predictive accuracy for at-risk versus non-at-risk classes.",{"name":70,"@type":61,"acceptedAnswer":71},"Which models are compared in the research?",{"text":72,"@type":64},"Ten models are compared, including random forest, decision tree, convolutional neural networks (CNN), multi-layer perceptron (MLP), and long short-term memory (LSTM).",{"name":74,"@type":61,"acceptedAnswer":75},"What is the key finding about deep learning performance?",{"text":76,"@type":64},"The DL-based predictive model shows superior performance in the higher-education setting, with the LSTM-based model achieving the highest and nearly perfect accuracy.","https://schema.org",{"og:url":32,"og:type":79,"og:title":10,"og:site_name":45,"og:description":12},"article",{"robots":81,"canonical":32},"index,follow",{"doc_id":83,"site_id":7},600513,1792217058,{"code":4,"msg":86,"data":87},"success",[88,92,96,100,105,109,114,118,123,126,130],{"id":22,"doc_module":4,"doc_module_name":25,"category_name":89,"show_sort_weight":90,"slug":91},"Story & Novel",90,"story-novel",{"id":26,"doc_module":4,"doc_module_name":25,"category_name":93,"show_sort_weight":94,"slug":95},"Literature",80,"literature",{"id":33,"doc_module":4,"doc_module_name":25,"category_name":97,"show_sort_weight":98,"slug":99},"Exam",70,"exam",{"id":101,"doc_module":4,"doc_module_name":25,"category_name":102,"show_sort_weight":103,"slug":104},5,"Comic",60,"comic",{"id":55,"doc_module":4,"doc_module_name":25,"category_name":106,"show_sort_weight":107,"slug":108},"Technology",50,"technology",{"id":110,"doc_module":4,"doc_module_name":25,"category_name":111,"show_sort_weight":112,"slug":113},7,"Healthcare",40,"healthcare",{"id":115,"doc_module":4,"doc_module_name":25,"category_name":29,"show_sort_weight":116,"slug":117},8,30,"research-report",{"id":119,"doc_module":4,"doc_module_name":25,"category_name":120,"show_sort_weight":121,"slug":122},9,"Religion & Spirituality",20,"religion-spirituality",{"id":121,"doc_module":4,"doc_module_name":25,"category_name":124,"show_sort_weight":121,"slug":125},"World Cup","world-cup",{"id":127,"doc_module":4,"doc_module_name":25,"category_name":128,"show_sort_weight":127,"slug":129},10,"Lifestyle","lifestyle",{"id":131,"doc_module":4,"doc_module_name":25,"category_name":132,"show_sort_weight":101,"slug":133},19,"General","general",{"code":4,"msg":86,"data":135},{"doc_id":83,"user_id":136,"nickname":42,"user_avatar":137,"doc_module":4,"category_id":115,"category_name":29,"doc_title":10,"doc_description":12,"doc_content":138,"file_id":139,"file_url":140,"file_type":141,"file_size":142,"view_count":55,"is_deleted":4,"is_public":22,"is_downloadable":22,"audit_status":22,"page_count":143,"language":144,"language_code":8,"site_id":7,"html_lang":8,"table_of_contents":145,"faqs":146,"seo_title":147,"seo_description":12,"update_tm":148,"read_time":112},962090893581,"https://ap-avatar.wpscdn.com/davatar_276721f389ce27ea32af1340a28f341c","OPEN ACCESS  \nResearch Article  \nContemporary Educational Technology, 2025, 17(4), ep606 ISSN: 1309-517X (Online)  \nEnhanced predictive performance: A comparative analysis of MLand DL models using on augmented LMS interaction data  \nKiran Fahd 1  \n 0009-0002-8213-3984  \nShah J. Miah 1*  \n 0000-0002-3783-8769  \n1 Newcastle Business School, College of Human and Social Futures, The University of Newcastle, Callaghan, NSW, AUSTRALIA * Corresponding author: [shah. miah@newcastle.edu.au](shah. miah@newcastle.edu.au)  \nCitation: Fahd, K., & Miah, S. J. (2025) . Enhanced predictive performance: A comparative analysis of ML and DL models using on augmented LMS interaction data. Contemporary Educational Technology, 17(4), ep606 .  \n[https://doi.org/10.30935/cedtech/17453](https://doi.org/10.30935/cedtech/17453)  \nARTICLE INFO ABSTRACT  \n\n| Received: 27 Feb 2025\u003Cbr>Accepted: 3 Jul 2025\u003Cbr>INTRODUCTION | The increasing reliance on the learning management system (LMS) in this era of digital education offers a vital source of student data that can be leveraged to predict student academic progress. Predicting student academic progress in higher education (HE) supports timely intervention and enhances student retention. This study develops and compares multiple machine learning (ML) and deep learning (DL) models to identify at-risk students based on students’ interaction data with LMS by leveraging an integrated DSR methodology. Multiple predictive models are developed by incorporating data augmentation and balancing techniques to address class imbalance and enhance the accuracy of the predictive model. The study compares ten different models to achieve the highest classification accuracy in predicting students at risk of failing through the integration of through the integration of both ML and DL algorithms, including random forest, decision tree, convolutional neural networks, multi-layer perceptron, and long short-term memory (LSTM) . The comparison results unscored the value of the DL based predictive model in the HE setting to precisely predict student academic performance, particularly the LTSM based model, which has the highest and nearly perfect accuracy. The existing LMS systems can incorporate this DL based predictive model to provide educational stakeholders with benefits and insights that support students’ academic journeys and institutional success.\u003Cbr>Keywords: machine learning, deep learning, design research, higher education, LMS data, student academic progress |\n| --- | --- |\n\nLearning management system (LMS) data provides a strong foundation for predicting student performance, a growing research area. LMS data is automatically generated through student behaviors and interactions with LMS that can be harnessed to anticipate their academic outcomes and implement appropriate support mechanisms if required. Early recognition of students at risk of failing allows for implementing appropriate intervention before students reach critical points of failure. Automated and datadriven approaches are increasingly being utilized in addition to traditional methods of identifying at-risk students, which often involve manual assessments, late detection, and delayed interventions (Tamada et al., 2021) . Early and timely identification of at-risk students using data-driven information systems (IS) can support education providers in taking appropriate measures effectively to enhance student progress and prevent academic procrastination, thereby improving the overall effectiveness of the educational program.  \nFigure 1. Overview of proposed educational technology (Source: Authors’ own elaboration)  \nIn recent literature, machine learning (ML) techniques have been used to analyze and detect patterns that can be transformed from educational data. Deep learning (DL) techniques, a subset of ML, employ neural networks (NN) for complex insight into datasets. Advanced data analysis techniques, such as ML and DL algorithms, are applied to LMS data to re","cbCaitMNO3JmFkQe","https://ap.wps.com/l/cbCaitMNO3JmFkQe","pdf",584935,16,"English","# Introduction\n## Problem and motivation: predicting at-risk students from LMS data\n## Modeling approach: ML and DL comparison\n# Methods and datasets\n## Data augmentation and class balancing\n# Results and implications\n## DL superiority in at-risk prediction","[{\"question\":\"What is the primary goal of the study?\",\"answer\":\"To develop and compare ML and DL models that use augmented LMS interaction data to identify higher-education students at risk of failing with high classification accuracy.\"},{\"question\":\"How does the study handle class imbalance in the predictive models?\",\"answer\":\"It incorporates data augmentation and balancing techniques as part of an integrated DSR methodology to improve predictive accuracy for at-risk versus non-at-risk classes.\"},{\"question\":\"Which models are compared in the research?\",\"answer\":\"Ten models are compared, including random forest, decision tree, convolutional neural networks (CNN), multi-layer perceptron (MLP), and long short-term memory (LSTM).\"},{\"question\":\"What is the key finding about deep learning performance?\",\"answer\":\"The DL-based predictive model shows superior performance in the higher-education setting, with the LSTM-based model achieving the highest and nearly perfect accuracy.\"}]","Enhanced predictive performance: A comparative analysis of ML and DL models using on augmented LMS interaction data | PDF",1792090842]