| Summary: | The surge in online education prompted by the COVID-19 pandemic outbreak has reshaped the traditional learning landscapes worldwide. With many educational institutions embracing online learning systems like Learning Management Systems (LMS) and Blackboard, a wealth of data is now available, offering insights into student academic progress. However, the transition to online distance learning has posed challenges, particularly in identifying struggling students, referred to as at-risk students, due to the absence of direct
visual and interpersonal cues.
To address this challenge, predictive analytics has emerged as a promising solution. This project leverages upon predictive analytics, specifically Random Forest, to predict at-risk students using the Open University Learning Analytics dataset as a case study. By analysing student behaviours and their interactions with the virtual learning environment, the model intends to identify students who may require additional support.
Furthermore, the project develops a Tableau dashboard to visualize the prediction results, providing educators with an explanatory tool to interpret and understand the predicted outcomes and provide tailored support to these outcomes. This report elucidates the methodology and development process of both the predictive model and the accompanying dashboard, offering insights into their implementation and potential impact on student support strategies in online education settings
|
|---|