What is needed to build a personalized recommender system for K-12 students’ E-Learning? Recommendations for future systems and a conceptual framework
/

Crafting personalised e-learning recommender systems for K–12 students

What is needed to build a personalized recommender system for K-12 students’ E-Learning? Recommendations for future systems and a conceptual framework.

Since the onset of the COVID-19 pandemic in 2020, e-learning has evolved into an indispensable aspect of education, with schools and colleges worldwide shifting classes online due to lockdowns, impacting over 1.6 billion students across 190 countries. This transformation has underscored the critical role of technology in ensuring educational accessibility, particularly in challenging times. Advancements in educational technology have facilitated the development of personalized learning plans tailored to each student’s unique preferences and learning styles, transitioning the vision of individualized education into a tangible reality. By leveraging smart systems, learning becomes not only more engaging but also more effective, prioritizing the fulfillment of every student’s educational needs and preventing anyone from being left behind or disengaged.

Amidst the wealth of online resources, students often find themselves overwhelmed, struggling to discern which materials to prioritize for their studies. This article is about a study titled “What is needed to build a personalized recommender system for K-12 students’ E-Learning? Recommendations for future systems and a conceptual framework”. By implementing personalized recommender systems, akin to having a personal tutor accessible through a computer, learning not only becomes more enjoyable but also more effective, tailored to each student’s individual needs, thus ensuring that no one is left behind or disinterested. In an educational landscape thirsting for innovation, personalized e-learning emerges as a beacon of hope, promising a revolution in K–12 education.

Credit. Author

How can personalized e-learning systems cater to K-12 students’ unique needs, enhancing engagement and educational outcomes effectively?

Jamallah M. Zawia

The heart of personalization: A four-stage framework

At the central of this study lies a meticulously structured framework comprising four essential stages: student profiling, material collection, material filtering, and validation. This framework transcends theoretical abstraction, serving as a pragmatic roadmap for developing e-learning systems to provide personalized learning experiences. With technology, educators can now curate learning environments tailored to each student, nurturing heightened engagement and a comprehensive grasp of the subject matter.

After a thorough five-year investigation, this study delved into a comprehensive review to isolate the fundamental elements of personalized recommendation systems tailored explicitly for educational environments. This extensive analysis sought to pinpoint crucial modules and personalization features essential for the efficacy of such systems. The culmination is a strategic blueprint that underscores primary components, personalization strategies, and evaluation methodologies. Furthermore, it introduces a conceptual framework designed to steer the development of these systems, guaranteeing their alignment with the distinctive requirements of educational settings while enriching the learning journey for students.

Figure 1. General Architecture of the proposed system
Credit. Author

Transforming challenges into opportunities

Pursuing personalized learning systems has revealed various methodologies and areas of focus within e-learning. Despite numerous studies exploring personalization, a noticeable void persists in meeting the distinct needs of K–12 students. This gap underscores the necessity for dedicated research to craft personalized e-learning systems tailored specifically to the individual requirements of school students. Given the disparities in learning environments, preferences, and teacher interactions between K–12 and higher education, there is a pressing need to address these unique needs to enrich the learning experience and outcomes of school students.

Implications for tomorrow’s classrooms

The study’s methodology involved a systematic literature review (SLR), adhering to Kitchenham’s protocols to mitigate bias and uphold relevance. It commenced by formulating research questions and identifying digital libraries for retrieving relevant studies. Employing stringent inclusion and exclusion criteria, the initial pool of 780 articles was meticulously narrowed to 23 pertinent ones. Subsequently, these articles underwent rigorous quality assessment, scrutinizing their clarity of objectives, methodology, contributions, data collection methodologies, and system-building processes. This rigorous approach aimed to identify critical components and features necessary for personalized e-learning systems, particularly focusing on K–12 educational settings.

The implications of this research extend far beyond the confines of academic discourse, providing a preview of a future where education is not only more engaging and effective but also equitable. As we stand on the brink of this educational revolution, policymakers, educators, and technology developers must collaborate to embrace these innovations. This collective effort is essential to ensuring that the potential of personalized e-learning materializes into a tangible reality for students across the globe.

A call to action

The study’s findings introduce a structured analysis of papers that discuss the development and effectiveness of personalized recommendation systems in e-learning. It dissects the pivotal elements and personalization attributes that address the distinctive requirements of students, with a primary focus on enriching their educational journey. Recommendations for future systems underscore the necessity for refined student profiling techniques and sophisticated material processing methodologies to guarantee that the resources dispensed are genuinely customized to each student’s unique learning style, capabilities, and preferences.

As we navigate towards this auspicious educational horizon, it becomes evident that the path forward demands more than just technological advancements—it necessitates a collaborative endeavor to reimagine and revamp the educational landscape. This article resonates as a rallying cry to all stakeholders within the educational sphere, urging them to embrace the transformative power of personalized e-learning wholeheartedly. By doing so, we can ensure that every student is empowered with the tools and opportunities needed to unlock their full potential.

🔬🧫🧪🔍🤓👩‍🔬🦠🔭📚

Journal reference

Zayet, T. M., Ismail, M. A., Almadi, S. H., Zawia, J. M. H., & Mohamad Nor, A. (2023). What is needed to build a personalized recommender system for K-12 students’ E-Learning? Recommendations for future systems and a conceptual framework. Education and information technologies28(6), 7487-7508. https://doi.org/10.1007/s10639-022-11489-4

Jamallah M. Zawia, an aspiring scholar and a PhD student at the University of Malaya, Malaysia, specialising in the development of recommender systems using deep learning for e-learning applications.

Maizatul Akmar Ismail, a distinguished scholar at Universiti Malaya, has made significant contributions across a variety of domains including Computer Science, Engineering, Telecommunications, Information Science & Library Science, and Business & Economics. Known in academic circles by various published names such as Maizatul Akmar Ismail, Maizatul Akmar Binti Ismail, and Maizatul Ismaili, Dr. Ismail's research has been instrumental in advancing our understanding of these fields. Recognised by the Web of Science with the ResearcherID B-8922-2010, Dr. Ismail's work is also associated with an ORCID iD (https://orcid.org/0000-0003-1877-7128), ensuring a durable link to their scholarly achievements.