Scholarlink

E-learning platforms have made a slow albeit steady penetration into the educational ecosystems over the years. Earlier, the scope of this phenomenon was limited primarily at the tertiary education level; however, the COVID-19 global pandemic and subsequent lockdowns necessitated the mainstreaming of e-learning tools at secondary and primary education levels as well. Post-pandemic e-learning tools retained their usability as education ecosystems transitioned toward a hybrid instruction model. E-learning has evolved significantly from the heydays of the pre-Covid era, but it can still be categorized as being in the early-mid-evolution phase. Currently, most of the platforms are neither intuitive nor provide any mechanism for empirically analyzing the students’ learning process. In this study, we aim to present an analytics tool incorporated within a proof-of-concept (POC) e-learning platform to showcase the capability and benefits of analytics in enhancing their learning experiences by monitoring and empirically analyzing student activities and to increase students’ and the instructor’s engagement in the overall learning process. Our POC e-learning platform has two main subsystems. The first subsystem exhaustively monitors the students’ activities in the learning platform and, based on their activity, classifies students as active or inactive. This subsystem uses multiple activity data points, including the number of logins into the platform, number of clicks, views of learning materials, and time spent on the platform for each enrolled course, among others, to create an analytics profile for every student. A K-nearest neighbor (KNN) approach is used to classify a student’s engagement in each enrolled course based on the monitored activities. The second subsystem employs the student analytics profile to recommend external learning videos from the YouTube website using a content-based recommender method to support weak learners and enhance their understanding of the topic they lack. Additionally, we integrated a reminder feature in the POC platform that reminds the students and instructors of upcoming due dates for assignments, final exam dates, and reminders for past submission dates. This reminder feature is focused mainly on the assessment events to give a reminder to students. The outcome of this study is a POC that can provide learning analytics and assist student monitoring throughout the course duration.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Learning Analytics Within a Proof-of-Concept E-Learning Platform for Students Performance Monitoring and Assistance

  • 章节

  • 发表日期：

    2025年

查看全文
  • Anusha Achuthan,
  • Thanish Natarajan,
  • Idrees Fazili

摘要

E-learning platforms have made a slow albeit steady penetration into the educational ecosystems over the years. Earlier, the scope of this phenomenon was limited primarily at the tertiary education level; however, the COVID-19 global pandemic and subsequent lockdowns necessitated the mainstreaming of e-learning tools at secondary and primary education levels as well. Post-pandemic e-learning tools retained their usability as education ecosystems transitioned toward a hybrid instruction model. E-learning has evolved significantly from the heydays of the pre-Covid era, but it can still be categorized as being in the early-mid-evolution phase. Currently, most of the platforms are neither intuitive nor provide any mechanism for empirically analyzing the students’ learning process. In this study, we aim to present an analytics tool incorporated within a proof-of-concept (POC) e-learning platform to showcase the capability and benefits of analytics in enhancing their learning experiences by monitoring and empirically analyzing student activities and to increase students’ and the instructor’s engagement in the overall learning process. Our POC e-learning platform has two main subsystems. The first subsystem exhaustively monitors the students’ activities in the learning platform and, based on their activity, classifies students as active or inactive. This subsystem uses multiple activity data points, including the number of logins into the platform, number of clicks, views of learning materials, and time spent on the platform for each enrolled course, among others, to create an analytics profile for every student. A K-nearest neighbor (KNN) approach is used to classify a student’s engagement in each enrolled course based on the monitored activities. The second subsystem employs the student analytics profile to recommend external learning videos from the YouTube website using a content-based recommender method to support weak learners and enhance their understanding of the topic they lack. Additionally, we integrated a reminder feature in the POC platform that reminds the students and instructors of upcoming due dates for assignments, final exam dates, and reminders for past submission dates. This reminder feature is focused mainly on the assessment events to give a reminder to students. The outcome of this study is a POC that can provide learning analytics and assist student monitoring throughout the course duration.