Designing Software Architecture for Digital Footprint Analysis in Educational Systems

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.