A model of recommendation service architecture for the formation of an individual educational trajectory for self-paced massive open online courses studying

Abstract

Data analytics tools in e-learning are widely used to improve the quality of massive open online courses (MOOCs), improve learner performance, and keep learners engaged in the learning process. However, the application of these tools and services is limited to the scope of a particular course. A recommendation service can help a learner to build an individual educational trajectory corresponding to their educational goals and existing knowledge and skills, to do it at their own pace, according to their own program and own schedule, and to use the variety of online courses of the whole educational platform. This tool is built based on a mathematical model using boolean programming.The used mathematical model is remarkable because it allows us to distinguish between the determination of an optimal set of MOOCs according to the given criteria and the construction of a schedule from this set. In this case, the optimal set of courses will be found only if it is possible to build such a sequence of courses so that the learner at each stage has all the knowledge and skills necessary to study the next course. This allows for a significant reduction in the amount of computation required.The article provides a list of prerequisites for creating the service, describes a possible architecture of the microservice approach and justifies its choice, presents an example of the used mathematical model, and evaluates the prospects of using the service.The proposed solution may be of practical interest to organizations that own e-learning platforms, as a means of increasing the proportion of students who successfully achieve their educational goals