A Withdrawal Prediction Model of At-Risk Learners Based on Behavioural Indicators

Nowadays, the virtual learning environment has become an ideal tool for professional self-development and bringing courses various learner audiences across world. There is currently increasing interest in researching topic of dropout low completion distance learning, with one main concerns being elevated rates occurrence. Therefore, early prediction withdrawal a major challenge, as well identifying factors, which contribute to this increasingly occurring phenomenon. In that regard, manuscript presents framework model data from The Open University, largest institutions. For purpose, we start by pre-processing dataset tackling challenge discretization process unbalanced data. Secondly, paper identifies semantical issues raw introducing new behavioural indicators. Finally, reckon on machine algorithms understand lack learners' commitment at stage.