Авторы:
Издательство:
Дата издания
01.09.2025На каком языке издана
ISSN
1991-346XОбласть исследования
Тип исследования
Аннотация (краткое описание)
Educational Recommender Systems (ERS) have become a critical component of modern digital learning environments, offering personalied learning pathways tailored to individual student needs, preferences, and behaviors. These systems utilie a wide range of data sourcesincluding demographic information, academic performance, cognitive characteristics, and behavioral patternsto recommend relevant courses, learning materials, and strategies that support student engagement and success. This paper presents a comprehensive review of current approaches used in the development of ERS, including collaborative filtering, content-based filtering, hybrid recommendation models, machine learning, deep learning algorithms, nowledge graphs, and eplainable artificial intelligence (XAI). Particular attention is given to the advantages and limitations of each approach, as well as ey challenges that persist in the implementation of ERS. These include the cold-start problem, data sparsity, scalability issues, lack of transparency in decision-maing, and concerns regarding user privacy and algorithmic fairness. The paper also eplores practical applications of ERS in higher education and large-scale online platforms such as MOOCs, where such systems have demonstrated positive impacts on learner motivation, retentionDOI
10.32014/2025.2518-1726.366Тип публикаций
Данный документ располагается в коллекциях
Публикации в КОКСОН
Файлы документа
Документ не загружен