Adaptive Multidimensional E-Learning Systems

Nadezhda Kunicina, Kamila Rakhimova, Jelena Caiko, Мадина Мансурова

2025

Closed access Metadata (CSL-JSON)
DOI 10.1109/aieee66149.2025.11050765
Data confidence 70 medium
How this score is calculated

It is built from checkable properties of the record itself: whether the article has a DOI, how many independent sources confirmed its data, whether the required fields are filled in, and how recent the latest confirmation is. Disagreement between sources and a missing identifier lower it. A value of 80 or above means the record was published without further checking; 60 to 79 means it is published at reduced confidence, with a manual check advisable. A record verified by an editor by hand is never pushed below 90 by a recalculation. This scores our data about the article, not the article itself.

Abstract

In recent years, adaptive learning systems have gained significant attention as a means to personalize education and enhance student outcomes. This study focuses on the development of a multidimensional adaptive e-learning system based on feedback mechanisms and data-driven decision-making. The proposed system aims to improve the learning experience by continuously analyzing student performance and adjusting educational content accordingly new data driven technologies [1].To achieve this, we employed a combination of data collection, processing, and analysis methods. The system gathers information from both primary and secondary sources. Primary data is obtained directly from students through surveys, tracking their demographic characteristics, learning preferences, and academic performance. These data points are used to build individualized learning profiles and identify deviations from expected progress [2]. The system also records user interactions, such as task completion rates, number of attempts, and error types, allowing for detailed behavioral analysis. Secondary data, including findings from authoritative research on adaptive learning, were used to refine the system’s decision-making algorithms.Machine learning techniques play a crucial role in data processing. We utilized the k-means clustering algorithm to categorize students [3] based on similar learning behaviors, such as average scores and time spent on tasks. This segmentation helps tailor instructional strategies to specific student groups [4], [5], thereby increasing engagement and effectiveness. The clustering approach was implemented using Python’s scikit-learn library, and the results were visualized to facilitate interpretation.The system was developed using modern web technologies: Flask for the backend, PostgreSQL for database management, and React.js for an interactive user interface. Data analysis was conducted in Jupyter Notebook using libraries such as pandas and scikit-learn.The practical significance of this research lies in its potential application in educational institutions to enhance teaching quality. By automating data analysis and generating personalized recommendations, the system reduces the burden on educators while improving student support. Additionally, the system can be expanded to analyze collaborative learning interactions and predict long-term academic performance.Future work will focus on integrating natural language processing (NLP) and neural network models to further refine personalization and feedback mechanisms.This study contributes to the advancement of adaptive learning technologies [6] by demonstrating a scalable and efficient approach to personalized education.