Tutor Impact Analysis Using Rough Set Theory in Personalized Learning Platforms

Personalized learning platforms have surfaced as a powerful resource for increasing student engagement and personalizing instruction. Tutors' actions are one of many factors that have been recognized as important in the mediation of learning and outcomes. Unfortunately, aforementioned research has struggled to evidently model the impact of the tutor due to overlapping attributes, ambiguous and uncertain learner behaviours, and imprecise evaluative criteria. As such, the reported conclusions and recommendations are restricted from data-driven understanding about the contribution of tutor interventions to learner movement. In response to this problem, the paper presents the Rough Set-Based Tutor Influence Evaluation Model (RS-TIEM) to systematically address uncertainty employing Rough Set Theory (RST) while identifying tutor features that are important in terms of student performance. RS-TIEM has used indiscernibility relations, created lower approximations and upper approximations, and formed decision rules to show certain and possible dependencies between tutor actions and the success of the learner. These tutor actions include factors such as feedback timeliness, explanation clarity, and responsive adjustability, which all evoke a positive contribution to learning outcome success, and thus should contribute to a description and subsequent RS-TIEM evaluation. Empirical evaluation has shown that RS-TIEM outperforms conventional methods in interpretive robustness, accuracy, and transparency, and ensures trustworthiness in monitoring tutor performance. The proposed model has been effectively utilized in personalized learning environments, enhancing tutor assignment and adaptive learning processes.

Перевод пока недоступен