Abstract:
Modern educational platforms and e-learning tools generate large volumes of
data about student activities, known as the "digital footprint". Analysis of this data allows
tracking student progress, identifying dropout risks, and predicting academic success. However,
many educational organizations lack the IT infrastructure necessary for comprehensive
collection, storage, and analysis of digital footprints. Existing solutions are often fragmented,
focused only on retrospective analysis, or require resource-intensive big data technologies,
making them difficult to implement in educational organizations.
This paper proposes a software system architecture for integrating and analyzing digital
footprints in educational systems that addresses these limitations. Approaches to data
collection through adapters, unification and storage of heterogeneous data, as well as
infrastructure solutions that ensure system reliability and scalability (microservice architecture,
containerization, CI/CD, monitoring) are described. An experimental analysis of the
effectiveness of the proposed approach is presented using the example of calculating
performance indicators, demonstrating a reduction in processing time and resource utilization
compared to traditional approaches. The proposed solutions allow for the creation of flexible
and high-performance educational analytics systems suitable for implementation even in
resource-constrained environments.
Key words and phrases:digital footprint, digital profile, software architecture, microservices, infrastructure, learning analytics.
Fulltext:
PDF file (1062 kB)