A systematic literature review on adaptive content recommenders in personalized learning environments from 2015 to 2020

Abstract

In personalized learning, each student gets a customized learning plan according to their pace of learning, instructional preferences, learning objects, etc. Hence the content recommender system in Personalized Learning Environment (PLE) should adapt to learner attributes and suggest appropriate learning resources to aid the learning process and improve the learning outcomes. This systematic literature review aims to analyze and summarize the studies on learning content recommenders in adaptive and personalized learning environments from 2015 to 2020. The publications were searched using proper keywords and filtered using the inclusion and exclusion criteria, which resulted in 52 publications. This paper summarizes the recent trends in research on different aspects of the recommender systems, such as learner attributes, recommendation methods, evaluation metrics, and the usability tests used by the researchers. It is observed that cognitive aspects of learners like learning style, preferences, knowledge level, etc., are used by most studies than non-cognitive aspects as social tags or trust. In most cases, recommendation engines are a hybrid of collaborative filtering, content-based filtering, ontological approaches, etc. All models were evaluated for the correctness of the prediction done, and a few studies have also done evaluations based on learner satisfaction or usability.

A Personalized Recommendation System for English Teaching Resources Based on Learning Behavior Detection

“…A residual mechanism is added in between each ST-GCN, and there are a total of nine ST-GCN units, each of which is composed of the GCN and TCN [33]. ese layers make up the TCN module and are referred to as the ReLU layer, dropout layer, 1-D convolution layer, and the batch normalization layer [34]. e following is how the spatial graph convolution for each node in the graph representing the human skeleton is calculated using the following equation:…”

Towards Personalized Adaptive Learning in e-Learning Recommender Systems

“…Then Javed et al [11] present review of a widely used methods in recommender systems, context-based and contentbased, and a hybrid method combining multiple methods in order to benefit of the advantage of each method to cover the disadvantages of each one. Just recently, Raj and Ranumol [12] provide a review of research papers on a period of time from 2015 to 2020, with critical study of adaptive recommender systems proposed comparing on one hand methods used in those systems, from the hybrid methods, content or agentbased, semantic web based, etc. On the other hand, they are also comparing the attributes such as the user content rating, learning style, knowledge level, etc.…”

Personalized Cognitive Counseling Process to Promote Digital Health

“…This is done by first determining the needs, interests and learning aspirations of each student teaching content and grading may vary according to the needs of the learner. A more or less tailored learning experience is then provided for each student (Raj et al, 2021), responding to their particular abilities, interests and needs. This is a practice that adjusts the pace and focus of teaching to meet the needs and goals of each learner.…”

“…(Stonerock & Blumenthal, 2017) In addition, it encourages the client to be able to find solutions to those problems on their own in a flexible manner, by determining their needs, interests, and inspirations with regard to learning, with an emphasis on flexible individual educational experiences (Lancaster & Stead, 2017) to meet the needs and goals of each client. (Raj et al, 2021) Personalized Cognitive Counseling consists of 5 steps as follows:1) Understand how learners learn best, 2) Designing a personal learning environment 3) Develop lessons 4) Choose and use tools, resources, and strategies 5) Evaluate learning…”